1 Change log file for Exim from version 4.21
2 -------------------------------------------
6 JH/01 Use SIZE on MAIL FROM in a cutthrough connection, if the destination
7 supports it and a size is available (ie. the sending peer gave us one).
9 JH/02 The obsolete acl condition "demime" is removed (finally, after ten
10 years of being deprecated). The replacements are the ACLs
11 acl_smtp_mime and acl_not_smtp_mime.
13 JH/03 Upgrade security requirements imposed for hosts_try_dane: previously
14 a downgraded non-dane trust-anchor for the TLS connection (CA-style)
15 or even an in-clear connection were permitted. Now, if the host lookup
16 was dnssec and dane was requested then the host is only used if the
17 TLSA lookup succeeds and is dnssec. Further hosts (eg. lower priority
18 MXs) will be tried (for hosts_try_dane though not for hosts_require_dane)
19 if one fails this test.
20 This means that a poorly-configured remote DNS will make it incommunicado;
21 but it protects against a DNS-interception attack on it.
23 JH/04 Bug 1810: make continued-use of an open smtp transport connection
24 non-noisy when a race steals the message being considered.
26 JH/05 If main configuration option tls_certificate is unset, generate a
27 selfsigned certificate for inbound TLS connections.
29 JH/06 Bug 165: hide more cases of password exposure - this time in expansions
30 in rewrites and routers.
32 JH/07 Retire gnutls_require_mac et.al. These were nonfunctional since 4.80
33 and logged a warning sing 4.83; now they are a configuration file error.
35 JH/08 Bug 1836: Fix crash in VRFY handling when handed an unqualified name
36 (lacking @domain). Apply the same qualification processing as RCPT.
38 JH/09 Bug 1804: Avoid writing msglog files when in -bh or -bhc mode.
40 JH/10 Support ${sha256:} applied to a string (as well as the previous
43 JH/11 Cutthrough: avoid using the callout hints db on a verify callout when
44 a cutthrough deliver is pending, as we always want to make a connection.
45 This also avoids re-routing the message when later placing the cutthrough
46 connection after a verify cache hit.
47 Do not update it with the verify result either.
49 JH/12 Cutthrough: disable when verify option success_on_redirect is used, and
50 when routing results in more than one destination address.
52 JH/13 Cutthrough: expand transport dkim_domain option when testing for dkim
53 signing (which inhibits the cutthrough capability). Previously only
54 the presence of an option was tested; now an expansion evaluating as
55 empty is permissible (obviously it should depend only on data available
56 when the cutthrough connection is made).
58 JH/14 Fix logging of errors under PIPELINING. Previously the log line giving
59 the relevant preceding SMTP command did not note the pipelining mode.
61 JH/15 Fix counting of empty lines in $body_linecount and $message_linecount.
62 Previously they were not counted.
64 JH/16 DANE: treat a TLSA lookup response having all non-TLSA RRs, the same
65 as one having no matching records. Previously we deferred the message
66 that needed the lookup.
68 JH/17 Fakereject: previously logged as a norml message arrival "<="; now
69 distinguished as "(=".
71 JH/18 Bug 1867: make the fail_defer_domains option on a dnslookup router work
72 for missing MX records. Previously it only worked for missing A records.
74 JH/19 Bug 1850: support Radius libraries that return REJECT_RC.
76 JH/20 Bug 1872: Ensure that acl_smtp_notquit is run when the connection drops
77 after the data-go-ahead and data-ack. Patch from Jason Betts.
79 JH/21 Bug 1846: Send DMARC forensic reports for reject and quaratine results,
80 even for a "none" policy. Patch from Tony Meyer.
82 JH/22 Fix continued use of a connection for further deliveries. If a port was
83 specified by a router, it must also match for the delivery to be
86 JH/23 Bug 1874: fix continued use of a connection for further deliveries.
87 When one of the recipients of a message was unsuitable for the connection
88 (has no matching addresses), we lost track of needing to mark it
89 deferred. As a result mail would be lost.
91 JH/24 Bug 1832: Log EHLO response on getting conn-close response for HELO.
93 JH/25 Decoding ACL controls is now done using a binary search; the sourcecode
94 takes up less space and should be simpler to maintain. Merge the ACL
95 condition decode tables also, with similar effect.
97 JH/26 Fix problem with one_time used on a redirect router which returned the
98 parent address unchanged. A retry would see the parent address marked as
99 delivered, so not attempt the (identical) child. As a result mail would
102 JH/27 Fix a possible security hole, wherein a process operating with the Exim
103 UID can gain a root shell. Credit to http://www.halfdog.net/ for
104 discovery and writeup. Ubuntu bug 1580454; no bug raised against Exim
107 JH/28 Enable {spool,log} filesystem space and inode checks as default.
108 Main config options check_{log,spool}_{inodes,space} are now
109 100 inodes, 10MB unless set otherwise in the configuration.
111 JH/29 Fix the connection_reject log selector to apply to the connect ACL.
112 Previously it only applied to the main-section connection policy
115 JH/30 Bug 1897: fix callouts connection fallback from TLS to cleartext.
120 JH/01 Bug 1664: Disable OCSP for GnuTLS library versions at/before 3.3.16
121 and 3.4.4 - once the server is enabled to respond to an OCSP request
122 it does even when not requested, resulting in a stapling non-aware
123 client dropping the TLS connection.
125 TF/01 Code cleanup: Overhaul the debug_selector and log_selector machinery to
126 support variable-length bit vectors. No functional change.
128 TF/02 Improve the consistency of logging incoming and outgoing interfaces.
129 The I= interface field on outgoing lines is now after the H= remote
130 host field, same as incoming lines. There is a separate
131 outgoing_interface log selector which allows you to disable the
134 JH/02 Bug 728: Close logfiles after a daemon-process "exceptional" log write.
135 If not running log_selector +smtp_connection the mainlog would be held
136 open indefinitely after a "too many connections" event, including to a
137 deleted file after a log rotate. Leave the per net connection logging
138 leaving it open for efficiency as that will be quickly detected by the
139 check on the next write.
141 HS/01 Bug 1671: Fix post transport crash.
142 Processing the wait-<transport> messages could crash the delivery
143 process if the message IDs didn't exist for some reason. When
144 using 'split_spool_directory=yes' the construction of the spool
145 file name failed already, exposing the same netto behaviour.
147 JH/03 Bug 425: Capture substrings in $regex1, $regex2 etc from regex &
148 mime_regex ACL conditions.
150 JH/04 Bug 1686: When compiled with EXPERIMENTAL_DSN_INFO: Add extra information
151 to DSN fail messages (bounces): remote IP, remote greeting, remote response
152 to HELO, local diagnostic string.
154 JH/05 Downgrade message for a TLS-certificate-based authentication fail from
155 log line to debug. Even when configured with a tls authenticator many
156 client connections are expected to not authenticate in this way, so
157 an authenticate fail is not an error.
159 HS/02 Add the Exim version string to the process info. This way exiwhat
160 gives some more detail about the running daemon.
162 JH/06 Bug 1395: time-limit cacheing of DNS lookups, to the TTL value. This may
163 matter for fast-change records such as DNSBLs.
165 JH/07 Bug 1678: Always record an interface option value, if set, as part of a
166 retry record, even if constant. There may be multiple transports with
167 different interface settings and the retry behaviour needs to be kept
170 JH/08 Bug 1586: exiqgrep now refuses to run if there are unexpected arguments.
172 JH/09 Bug 1700: ignore space & tab embedded in base64 during decode.
174 JH/10 Bug 840: fix log_defer_output option of pipe transport
176 JH/11 Bug 830: use same host for all RCPTS of a message, even under
177 hosts_randomize. This matters a lot when combined with mua_wrapper.
179 JH/12 Bug 1706: percent and underbar characters are no longer escaped by the
180 ${quote_pgsql:<string>} operator.
182 JH/13 Bug 1708: avoid misaligned access in cached lookup.
184 JH/14 Change header file name for freeradius-client. Relevant if compiling
185 with Radius support; from the Gentoo tree and checked under Fedora.
187 JH/15 Bug 1712: Introduce $prdr_requested flag variable
189 JH/16 Bug 1714: Permit an empty string as expansion result for transport
190 option transport_filter, meaning no filtering.
192 JH/17 Bug 1713: Fix non-PDKIM_DEBUG build. Patch from Jasen Betts.
194 JH/18 Bug 1709: When built with TLS support, the tls_advertise_hosts option now
195 defaults to "*" (all hosts). The variable is now available when not built
196 with TLS, default unset, mainly to enable keeping the testuite sane.
197 If a server certificate is not supplied (via tls_certificate) an error is
198 logged, and clients will find TLS connections fail on startup. Presumably
199 they will retry in-clear.
200 Packagers of Exim are strongly encouraged to create a server certificate
201 at installation time.
203 HS/03 Add -bP config_file as a synonym for -bP configure_file, for consistency
204 with the $config_file variable.
206 JH/19 Two additional event types: msg:rcpt:defer and msg:rcpt:host:defer. Both
207 in transport context, after the attempt, and per-recipient. The latter type
208 is per host attempted. The event data is the error message, and the errno
209 information encodes the lookup type (A vs. MX) used for the (first) host,
210 and the trailing two digits of the smtp 4xx reponse.
212 GF/01 Bug 1715: Fix for race condition in exicyclog, where exim could attempt
213 to write to mainlog (or rejectlog, paniclog) in the window between file
214 creation and permissions/ownership being changed. Particularly affects
215 installations where exicyclog is run as root, rather than exim user;
216 result is that the running daemon panics and dies.
218 JH/20 Bug 1701: For MySQL lookups, support MySQL config file option group names.
220 JH/21 Bug 1720: Add support for priority groups and weighted-random proxy
221 selection for the EXPERIMENTAL_SOCKS feature, via new per-proxy options
222 "pri" and "weight". Note that the previous implicit priority given by the
223 list order is no longer honoured.
225 JH/22 Bugs 963, 1721: Fix some corner cases in message body canonicalisation
228 JH/23 Move SOCKS5 support from Experimental to mainline, enabled for a build
229 by defining SUPPORT_SOCKS.
231 JH/26 Move PROXY support from Experimental to mainline, enabled for a build
232 by defining SUPPORT_PROXY. Note that the proxy_required_hosts option
233 is renamed to hosts_proxy, and the proxy_{host,target}_{address,port}.
234 variables are renamed to proxy_{local,external}_{address,port}.
236 JH/27 Move Internationalisation support from Experimental to mainline, enabled
237 for a build by defining SUPPORT_I18N
239 JH/28 Bug 1745: Fix redis lookups to handle (quoted) spaces embedded in parts
240 of the query string, and make ${quote_redis:} do that quoting.
242 JH/29 Move Events support from Experimental to mainline, enabled by default
243 and removable for a build by defining DISABLE_EVENT.
245 JH/30 Updated DANE implementation code to current from Viktor Dukhovni.
247 JH/31 Fix bug with hosts_connection_nolog and named-lists which were wrongly
248 cached by the daemon.
250 JH/32 Move Redis support from Experimental to mainline, enabled for a build
251 by defining LOOKUP_REDIS. The libhiredis library is required.
253 JH/33 Bug 1748: Permit ACL dnslists= condition in non-smtp ACLs if explicit
254 keys are given for lookup.
256 JH/34 Bug 1192: replace the embedded copy of PolarSSL RSA routines in the DKIM
257 support, by using OpenSSL or GnuTLS library ones. This means DKIM is
258 only supported when built with TLS support. The PolarSSL SHA routines
259 are still used when the TLS library is too old for convenient support.
261 JH/35 Require SINGLE_DH_USE by default in OpenSSL (main config option
262 openssl_options), for security. OpenSSL forces this from version 1.1.0
263 server-side so match that on older versions.
265 JH/36 Bug 1778: longstanding bug in memory use by the ${run } expansion: A fresh
266 allocation for $value could be released as the expansion processing
267 concluded, but leaving the global pointer active for it.
269 JH/37 Bug 1769: Permit a VRFY ACL to override the default 252 response,
270 and to use the domains and local_parts ACL conditions.
272 JH/38 Fix cutthrough bug with body lines having a single dot. The dot was
273 incorrectly not doubled on cutthrough transmission, hence seen as a
274 body-termination at the receiving system - resulting in truncated mails.
275 Commonly the sender saw a TCP-level error, and retransmitted the message
276 via the normal store-and-forward channel. This could result in duplicates
277 received - but deduplicating mailstores were liable to retain only the
278 initial truncated version.
280 JH/39 Bug 1781: Fix use of DKIM private-keys having trailing '=' in the base-64.
282 JH/40 Fix crash in queryprogram router when compiled with EXPERIMENTAL_SRS.
284 JH/41 Bug 1792: Fix selection of headers to sign for DKIM: bottom-up. While
285 we're in there, support oversigning also; bug 1309.
287 JH/42 Bug 1796: Fix error logged on a malware scanner connection failure.
289 HS/04 Add support for keep_environment and add_environment options.
291 JH/43 Tidy coding issues detected by gcc --fsanitize=undefined. Some remain;
292 either intentional arithmetic overflow during PRNG, or testing config-
295 JH/44 Bug 1800: The combination of a -bhc commandline option and cutthrough
296 delivery resulted in actual delivery. Cancel cutthrough before DATA
299 JH/45 Fix cutthrough, when connection not opened by verify and target hard-
300 rejects a recipient: pass the reject to the originator.
302 JH/46 Multiple issues raised by Coverity. Some were obvious or plausible bugs.
303 Many were false-positives and ignorable, but it's worth fixing the
306 JH/47 Fix build on HP-UX and older Solaris, which need (un)setenv now also
307 for the new environment-manipulation done at startup. Move the routines
308 from being local to tls.c to being global via the os.c file.
310 JH/48 Bug 1807: Fix ${extract } for the numeric/3-string case. While preparsing
311 an extract embedded as result-arg for a map, the first arg for extract
312 is unavailable so we cannot tell if this is a numbered or keyed
313 extraction. Accept either.
319 JH/01 Bug 1545: The smtp transport option "retry_include_ip_address" is now
322 JH/02 The smtp transport option "multi_domain" is now expanded.
324 JH/03 The smtp transport now requests PRDR by default, if the server offers
327 JH/04 Certificate name checking on server certificates, when exim is a client,
328 is now done by default. The transport option tls_verify_cert_hostnames
329 can be used to disable this per-host. The build option
330 EXPERIMENTAL_CERTNAMES is withdrawn.
332 JH/05 The value of the tls_verify_certificates smtp transport and main options
333 default to the word "system" to access the system default CA bundle.
334 For GnuTLS, only version 3.0.20 or later.
336 JH/06 Verification of the server certificate for a TLS connection is now tried
337 (but not required) by default. The verification status is now logged by
338 default, for both outbound TLS and client-certificate supplying inbound
341 JH/07 Changed the default rfc1413 lookup settings to disable calls. Few
344 JH/08 The EXPERIMENTAL_DSN compile option is no longer needed; all Delivery
345 Status Notification (bounce) messages are now MIME format per RFC 3464.
346 Support for RFC 3461 DSN options NOTIFY,ENVID,RET,ORCPT can be advertised
347 under the control of the dsn_advertise_hosts option, and routers may
348 have a dsn_lasthop option.
350 JH/09 A timeout of 2 minutes is now applied to all malware scanner types by
351 default, modifiable by a malware= option. The list separator for
352 the options can now be changed in the usual way. Bug 68.
354 JH/10 The smtp_receive_timeout main option is now expanded before use.
356 JH/11 The incoming_interface log option now also enables logging of the
357 local interface on delivery outgoing connections.
359 JH/12 The cutthrough-routing facility now supports multi-recipient mails,
360 if the interface and destination host and port all match.
362 JH/13 Bug 344: The verify = reverse_host_lookup ACL condition now accepts a
365 JH/14 Bug 1573: The spam= ACL condition now additionally supports Rspamd.
366 Patch from Andrew Lewis.
368 JH/15 Bug 670: The spamd_address main option (for the spam= ACL condition)
369 now supports optional time-restrictions, weighting, and priority
370 modifiers per server. Patch originally by <rommer@active.by>.
372 JH/16 The spamd_address main option now supports a mixed list of local
373 and remote servers. Remote servers can be IPv6 addresses, and
374 specify a port-range.
376 JH/17 Bug 68: The spamd_address main option now supports an optional
377 timeout value per server.
379 JH/18 Bug 1581: Router and transport options headers_add/remove can
380 now have the list separator specified.
382 JH/19 Bug 392: spamd_address, and clamd av_scanner, now support retry
385 JH/20 Bug 1571: Ensure that $tls_in_peerdn is set, when verification fails
388 JH/21 Support for the A6 type of dns record is withdrawn.
390 JH/22 Bug 608: The result of a QUIT or not-QUIT toplevel ACL now matters
391 rather than the verbs used.
393 JH/23 Bug 1572: Increase limit on SMTP confirmation message copy size
394 from 255 to 1024 chars.
396 JH/24 Verification callouts now attempt to use TLS by default.
398 HS/01 DNSSEC options (dnssec_require_domains, dnssec_request_domains)
399 are generic router options now. The defaults didn't change.
401 JH/25 Bug 466: Add RFC2322 support for MIME attachment filenames.
402 Original patch from Alexander Shikoff, worked over by JH.
404 HS/02 Bug 1575: exigrep falls back to autodetection of compressed
405 files if ZCAT_COMMAND is not executable.
407 JH/26 Bug 1539: Add timout/retry options on dnsdb lookups.
409 JH/27 Bug 286: Support SOA lookup in dnsdb lookups.
411 JH/28 Bug 1588: Do not use the A lookup following an AAAA for setting the FQDN.
412 Normally benign, it bites when the pair was led to by a CNAME;
413 modern usage is to not canoicalize the domain to a CNAME target
414 (and we were inconsistent anyway for A-only vs AAAA+A).
416 JH/29 Bug 1632: Removed the word "rejected" from line logged for ACL discards.
418 JH/30 Check the forward DNS lookup for DNSSEC, in addition to the reverse,
419 when evaluating $sender_host_dnssec.
421 JH/31 Check the HELO verification lookup for DNSSEC, adding new
422 $sender_helo_dnssec variable.
424 JH/32 Bug 1397: Enable ECDHE on OpenSSL, just the NIST P-256 curve.
426 JH/33 Bug 1346: Note MAIL cmd seen in -bS batch, to avoid smtp_no_mail log.
428 JH/34 Bug 1648: Fix a memory leak seen with "mailq" and large queues.
430 JH/35 Bug 1642: Fix support of $spam_ variables at delivery time. Was
431 documented as working, but never had. Support all but $spam_report.
433 JH/36 Bug 1659: Guard checking of input smtp commands again pseudo-command
434 added for tls authenticator.
436 HS/03 Add perl_taintmode main config option
441 TL/01 When running the test suite, the README says that variables such as
442 no_msglog_check are global and can be placed anywhere in a specific
443 test's script, however it was observed that placement needed to be near
444 the beginning for it to behave that way. Changed the runtest perl
445 script to read through the entire script once to detect and set these
446 variables, reset to the beginning of the script, and then run through
447 the script parsing/test process like normal.
449 TL/02 The BSD's have an arc4random API. One of the functions to induce
450 adding randomness was arc4random_stir(), but it has been removed in
451 OpenBSD 5.5. Detect this OpenBSD version and skip calling this
452 function when detected.
454 JH/01 Expand the EXPERIMENTAL_TPDA feature. Several different events now
455 cause callback expansion.
457 TL/03 Bugzilla 1518: Clarify "condition" processing in routers; that
458 syntax errors in an expansion can be treated as a string instead of
459 logging or causing an error, due to the internal use of bool_lax
460 instead of bool when processing it.
462 JH/02 Add EXPERIMENTAL_DANE, allowing for using the DNS as trust-anchor for
463 server certificates when making smtp deliveries.
465 JH/03 Support secondary-separator specifier for MX, SRV, TLSA lookups.
467 JH/04 Add ${sort {list}{condition}{extractor}} expansion item.
469 TL/04 Bugzilla 1216: Add -M (related messages) option to exigrep.
471 TL/05 GitHub Issue 18: Adjust logic testing for true/false in redis lookups.
472 Merged patch from Sebastian Wiedenroth.
474 JH/05 Fix results-pipe from transport process. Several recipients, combined
475 with certificate use, exposed issues where response data items split
476 over buffer boundaries were not parsed properly. This eventually
477 resulted in duplicates being sent. This issue only became common enough
478 to notice due to the introduction of conection certificate information,
479 the item size being so much larger. Found and fixed by Wolfgang Breyha.
481 JH/06 Bug 1533: Fix truncation of items in headers_remove lists. A fixed
482 size buffer was used, resulting in syntax errors when an expansion
485 JH/07 Add support for directories of certificates when compiled with a GnuTLS
486 version 3.3.6 or later.
488 JH/08 Rename the TPDA expermimental facility to Event Actions. The #ifdef
489 is EXPERIMENTAL_EVENT, the main-configuration and transport options
490 both become "event_action", the variables become $event_name, $event_data
491 and $event_defer_errno. There is a new variable $verify_mode, usable in
492 routers, transports and related events. The tls:cert event is now also
493 raised for inbound connections, if the main configuration event_action
496 TL/06 In test suite, disable OCSP for old versions of openssl which contained
497 early OCSP support, but no stapling (appears to be less than 1.0.0).
499 JH/09 When compiled with OpenSSL and EXPERIMENTAL_CERTNAMES, the checks on
500 server certificate names available under the smtp transport option
501 "tls_verify_cert_hostname" now do not permit multi-component wildcard
504 JH/10 Time-related extraction expansions from certificates now use the main
505 option "timezone" setting for output formatting, and are consistent
506 between OpenSSL and GnuTLS compilations. Bug 1541.
508 JH/11 Fix a crash in mime ACL when meeting a zero-length, quoted or RFC2047-
509 encoded parameter in the incoming message. Bug 1558.
511 JH/12 Bug 1527: Autogrow buffer used in reading spool files. Since they now
512 include certificate info, eximon was claiming there were spoolfile
515 JH/13 Bug 1521: Fix ldap lookup for single-attr request, multiple-attr return.
517 JH/14 Log delivery-related information more consistently, using the sequence
518 "H=<name> [<ip>]" wherever possible.
520 TL/07 Bug 1547: Omit RFCs from release. Draft and RFCs have licenses which
521 are problematic for Debian distribution, omit them from the release
524 JH/15 Updates and fixes to the EXPERIMENTAL_DSN feature.
526 JH/16 Fix string representation of time values on 64bit time_t anchitectures.
529 JH/17 Fix a null-indirection in certextract expansions when a nondefault
530 output list separator was used.
535 TL/01 Bugzilla 1506: Re-add a 'return NULL' to silence complaints from static
536 checkers that were complaining about end of non-void function with no
539 JH/01 Bug 1513: Fix parsing of quoted parameter values in MIME headers.
540 This was a regression intruduced in 4.83 by another bugfix.
542 JH/02 Fix broken compilation when EXPERIMENTAL_DSN is enabled.
544 TL/02 Bug 1509: Fix exipick for enhanced spoolfile specification used when
545 EXPERIMENTAL_DSN is enabled. Fix from Wolfgang Breyha.
551 TF/01 Correctly close the server side of TLS when forking for delivery.
553 When a message was received over SMTP with TLS, Exim failed to clear up
554 the incoming connection properly after forking off the child process to
555 deliver the message. In some situations the subsequent outgoing
556 delivery connection happened to have the same fd number as the incoming
557 connection previously had. Exim would try to use TLS and fail, logging
558 a "Bad file descriptor" error.
560 TF/02 Portability fix for building lookup modules on Solaris when the xpg4
561 utilities have not been installed.
563 JH/01 Fix memory-handling in use of acl as a conditional; avoid free of
564 temporary space as the ACL may create new global variables.
566 TL/01 LDAP support uses per connection or global context settings, depending
567 upon the detected version of the libraries at build time.
569 TL/02 Experimental Proxy Protocol support: allows a proxied SMTP connection
570 to extract and use the src ip:port in logging and expansions as if it
571 were a direct connection from the outside internet. PPv2 support was
572 updated based on HAProxy spec change in May 2014.
574 JH/02 Add ${listextract {number}{list}{success}{fail}}.
576 TL/03 Bugzilla 1433: Fix DMARC SEGV with specific From header contents.
577 Properly escape header and check for NULL return.
579 PP/01 Continue incomplete 4.82 PP/19 by fixing docs too: use dns_dnssec_ok
582 JH/03 Bugzilla 1157: support log_selector smtp_confirmation for lmtp.
584 TL/04 Add verify = header_names_ascii check to reject email with non-ASCII
585 characters in header names, implemented as a verify condition.
586 Contributed by Michael Fischer v. Mollard.
588 TL/05 Rename SPF condition results err_perm and err_temp to standardized
589 results permerror and temperror. Previous values are deprecated but
590 still accepted. In a future release, err_perm and err_temp will be
591 completely removed, which will be a backward incompatibility if the
592 ACL tests for either of these two old results. Patch contributed by
593 user bes-internal on the mailing list.
595 JH/04 Add ${utf8clean:} operator. Contributed by Alex Rau.
597 JH/05 Bugzilla 305: Log incoming-TLS details on rejects, subject to log
598 selectors, in both main and reject logs.
600 JH/06 Log outbound-TLS and port details, subject to log selectors, for a
603 JH/07 Add malware type "sock" for talking to simple daemon.
605 JH/08 Bugzilla 1371: Add tls_{,try_}verify_hosts to smtp transport.
607 JH/09 Bugzilla 1431: Support (with limitations) headers_add/headers_remove in
608 routers/transports under cutthrough routing.
610 JH/10 Bugzilla 1005: ACL "condition =" should accept values which are negative
611 numbers. Touch up "bool" conditional to keep the same definition.
613 TL/06 Remove duplicated language in spec file from 4.82 TL/16.
615 JH/11 Add dnsdb tlsa lookup. From Todd Lyons.
617 JH/12 Expand items in router/transport headers_add or headers_remove lists
618 individually rather than the list as a whole. Bug 1452.
620 Required for reasonable handling of multiple headers_ options when
621 they may be empty; requires that headers_remove items with embedded
622 colons must have them doubled (or the list-separator changed).
624 TL/07 Add new dmarc expansion variable $dmarc_domain_policy to directly
625 view the policy declared in the DMARC record. Currently, $dmarc_status
626 is a combined value of both the record presence and the result of the
629 JH/13 Fix handling of $tls_cipher et.al. in (non-verify) transport. Bug 1455.
631 JH/14 New options dnssec_request_domains, dnssec_require_domains on the
632 dnslookup router and the smtp transport (applying to the forward
635 TL/08 Bugzilla 1453: New LDAP "SERVERS=" option allows admin to override list
636 of ldap servers used for a specific lookup. Patch provided by Heiko
639 JH/18 New options dnssec_lax, dnssec_strict on dnsdb lookups.
640 New variable $lookup_dnssec_authenticated for observability.
642 TL/09 Bugzilla 609: Add -C option to exiqgrep, specify which exim.conf to use.
643 Patch submitted by Lars Timman.
645 JH/19 EXPERIMENTAL_OCSP support under GnuTLS. Bug 1459.
647 TL/10 Bugzilla 1454: New -oMm option to pass message reference to Exim.
648 Requires trusted mode and valid format message id, aborts otherwise.
649 Patch contributed by Heiko Schlichting.
651 JH/20 New expansion variables tls_(in,out)_(our,peer)cert, and expansion item
652 certextract with support for various fields. Bug 1358.
654 JH/21 Observability of OCSP via variables tls_(in,out)_ocsp. Stapling
655 is requested by default, modifiable by smtp transport option
658 JH/22 Expansion operators ${md5:string} and ${sha1:string} can now
659 operate on certificate variables to give certificate fingerprints
660 Also new ${sha256:cert_variable}.
662 JH/23 The PRDR feature is moved from being Experimental into the mainline.
664 TL/11 Bug 1119: fix memory allocation in string_printing2(). Patch from
665 Christian Aistleitner.
667 JH/24 The OCSP stapling feature is moved from Experimental into the mainline.
669 TL/12 Bug 1444: Fix improper \r\n sequence handling when writing spool
670 file. Patch from Wolfgang Breyha.
672 JH/25 Expand the coverage of the delivery $host and $host_address to
673 client authenticators run in verify callout. Bug 1476.
675 JH/26 Port service names are now accepted for tls_on_connect_ports, to
676 align with daemon_smtp_ports. Bug 72.
678 TF/03 Fix udpsend. The ip_connectedsocket() function's socket type
679 support and error reporting did not work properly.
681 TL/13 Bug 1495: Exiqgrep check if -C config file specified on cli exists
682 and is readable. Patch from Andrew Colin Kissa.
684 TL/14 Enhance documentation of ${run expansion and how it parses the
685 commandline after expansion, particularly in the case when an
686 unquoted variable expansion results in an empty value.
688 JH/27 The TLS SNI feature was broken in 4.82. Fix it.
690 PP/02 Fix internal collision of T_APL on systems which support RFC3123
691 by renaming away from it. Addresses GH issue 15, reported by
694 JH/28 Fix parsing of MIME headers for parameters with quoted semicolons.
696 TL/15 SECURITY: prevent double expansion in math comparison functions
697 (can expand unsanitized data). Not remotely exploitable.
704 PP/01 Add -bI: framework, and -bI:sieve for querying sieve capabilities.
706 PP/02 Make -n do something, by making it not do something.
707 When combined with -bP, the name of an option is not output.
709 PP/03 Added tls_dh_min_bits SMTP transport driver option, only honoured
712 PP/04 First step towards DNSSEC, provide $sender_host_dnssec for
713 $sender_host_name and config options to manage this, and basic check
716 PP/05 DSCP support for outbound connections and control modifier for inbound.
718 PP/06 Cyrus SASL: set local and remote IP;port properties for driver.
719 (Only plugin which currently uses this is kerberos4, which nobody should
720 be using, but we should make it available and other future plugins might
721 conceivably use it, even though it would break NAT; stuff *should* be
722 using channel bindings instead).
724 PP/07 Handle "exim -L <tag>" to indicate to use syslog with tag as the process
725 name; added for Sendmail compatibility; requires admin caller.
726 Handle -G as equivalent to "control = suppress_local_fixups" (we used to
727 just ignore it); requires trusted caller.
728 Also parse but ignore: -Ac -Am -X<logfile>
731 TL/01 Bugzilla 1258 - Refactor MAIL FROM optional args processing.
733 TL/02 Add +smtp_confirmation as a default logging option.
735 TL/03 Bugzilla 198 - Implement remove_header ACL modifier.
736 Patch by Magnus Holmgren from 2007-02-20.
738 TL/04 Bugzilla 1281 - Spec typo.
739 Bugzilla 1283 - Spec typo.
740 Bugzilla 1290 - Spec grammar fixes.
742 TL/05 Bugzilla 1285 - Spec omission, fix docbook errors for spec.txt creation.
744 TL/06 Add Experimental DMARC support using libopendmarc libraries.
746 TL/07 Fix an out of order global option causing a segfault. Reported to dev
747 mailing list by by Dmitry Isaikin.
749 JH/01 Bugzilla 1201 & 304 - New cutthrough-delivery feature, with TLS support.
751 JH/02 Support "G" suffix to numbers in ${if comparisons.
753 PP/08 Handle smtp transport tls_sni option forced-fail for OpenSSL.
755 NM/01 Bugzilla 1197 - Spec typo
756 Bugzilla 1196 - Spec examples corrections
758 JH/03 Add expansion operators ${listnamed:name} and ${listcount:string}
760 PP/09 Add gnutls_allow_auto_pkcs11 option (was originally called
761 gnutls_enable_pkcs11, but renamed to more accurately indicate its
764 PP/10 Let Linux makefile inherit CFLAGS/CFLAGS_DYNAMIC.
765 Pulled from Debian 30_dontoverridecflags.dpatch by Andreas Metzler.
767 JH/04 Add expansion item ${acl {name}{arg}...}, expansion condition
768 "acl {{name}{arg}...}", and optional args on acl condition
771 JH/05 Permit multiple router/transport headers_add/remove lines.
773 JH/06 Add dnsdb pseudo-lookup "a+" to do an "aaaa" + "a" combination.
775 JH/07 Avoid using a waiting database for a single-message-only transport.
776 Performance patch from Paul Fisher. Bugzilla 1262.
778 JH/08 Strip leading/trailing newlines from add_header ACL modifier data.
781 JH/09 Add $headers_added variable, with content from use of ACL modifier
782 add_header (but not yet added to the message). Bugzilla 199.
784 JH/10 Add 8bitmime log_selector, for 8bitmime status on the received line.
785 Pulled from Bugzilla 817 by Wolfgang Breyha.
787 PP/11 SECURITY: protect DKIM DNS decoding from remote exploit.
789 (nb: this is the same fix as in Exim 4.80.1)
791 JH/11 Add A= logging on delivery lines, and a client_set_id option on
794 JH/12 Add optional authenticated_sender logging to A= and a log_selector
797 PP/12 Unbreak server_set_id for NTLM/SPA auth, broken by 4.80 PP/29.
799 PP/13 Dovecot auth: log better reason to rejectlog if Dovecot did not
800 advertise SMTP AUTH mechanism to us, instead of a generic
801 protocol violation error. Also, make Exim more robust to bad
802 data from the Dovecot auth socket.
804 TF/01 Fix ultimate retry timeouts for intermittently deliverable recipients.
806 When a queue runner is handling a message, Exim first routes the
807 recipient addresses, during which it prunes them based on the retry
808 hints database. After that it attempts to deliver the message to
809 any remaining recipients. It then updates the hints database using
812 So if a recipient address works intermittently, it can get repeatedly
813 deferred at routing time. The retry hints record remains fresh so the
814 address never reaches the final cutoff time.
816 This is a fairly common occurrence when a user is bumping up against
817 their storage quota. Exim had some logic in its local delivery code
818 to deal with this. However it did not apply to per-recipient defers
819 in remote deliveries, e.g. over LMTP to a separate IMAP message store.
821 This change adds a proper retry rule check during routing so that the
822 final cutoff time is checked against the message's age. We only do
823 this check if there is an address retry record and there is not a
824 domain retry record; this implies that previous attempts to handle
825 the address had the retry_use_local_parts option turned on. We use
826 this as an approximation for the destination being like a local
827 delivery, as in LMTP.
829 I suspect this new check makes the old local delivery cutoff check
830 redundant, but I have not verified this so I left the code in place.
832 TF/02 Correct gecos expansion when From: is a prefix of the username.
834 Test 0254 submits a message to Exim with the header
838 When I ran the test suite under the user fanf2, Exim expanded
839 the header to contain my full name, whereas it should have added
840 a Resent-Sender: header. It erroneously treats any prefix of the
841 username as equal to the username.
843 This change corrects that bug.
845 GF/01 DCC debug and logging tidyup
846 Error conditions log to paniclog rather than rejectlog.
847 Debug lines prefixed by "DCC: " to remove any ambiguity.
849 TF/03 Avoid unnecessary rebuilds of lookup-related code.
851 PP/14 Fix OCSP reinitialisation in SNI handling for Exim/TLS as server.
852 Bug spotted by Jeremy Harris; was flawed since initial commit.
853 Would have resulted in OCSP responses post-SNI triggering an Exim
854 NULL dereference and crash.
856 JH/13 Add $router_name and $transport_name variables. Bugzilla 308.
858 PP/15 Define SIOCGIFCONF_GIVES_ADDR for GNU Hurd.
859 Bug detection, analysis and fix by Samuel Thibault.
860 Bugzilla 1331, Debian bug #698092.
862 SC/01 Update eximstats to watch out for senders sending 'HELO [IpAddr]'
864 JH/14 SMTP PRDR (http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt).
865 Server implementation by Todd Lyons, client by JH.
866 Only enabled when compiled with EXPERIMENTAL_PRDR. A new
867 config variable "prdr_enable" controls whether the server
868 advertises the facility. If the client requests PRDR a new
869 acl_data_smtp_prdr ACL is called once for each recipient, after
870 the body content is received and before the acl_smtp_data ACL.
871 The client is controlled by bolth of: a hosts_try_prdr option
872 on the smtp transport, and the server advertisement.
873 Default client logging of deliveries and rejections involving
874 PRDR are flagged with the string "PRDR".
876 PP/16 Fix problems caused by timeouts during quit ACLs trying to double
877 fclose(). Diagnosis by Todd Lyons.
879 PP/17 Update configure.default to handle IPv6 localhost better.
880 Patch by Alain Williams (plus minor tweaks).
883 PP/18 OpenSSL made graceful with empty tls_verify_certificates setting.
884 This is now consistent with GnuTLS, and is now documented: the
885 previous undocumented portable approach to treating the option as
886 unset was to force an expansion failure. That still works, and
887 an empty string is now equivalent.
889 PP/19 Renamed DNSSEC-enabling option to "dns_dnssec_ok", to make it
890 clearer that Exim is using the DO (DNSSEC OK) EDNS0 resolver flag,
891 not performing validation itself.
893 PP/20 Added force_command boolean option to pipe transport.
894 Patch from Nick Koston, of cPanel Inc.
896 JH/15 AUTH support on callouts (and hence cutthrough-deliveries).
899 TF/04 Added udpsend ACL modifer and hexquote expansion operator
901 PP/21 Fix eximon continuous updating with timestamped log-files.
902 Broken in a format-string cleanup in 4.80, missed when I repaired the
903 other false fix of the same issue.
904 Report and fix from Heiko Schlichting.
907 PP/22 Guard LDAP TLS usage against Solaris LDAP variant.
908 Report from Prashanth Katuri.
910 PP/23 Support safari_ecdhe_ecdsa_bug for openssl_options.
911 It's SecureTransport, so affects any MacOS clients which use the
912 system-integrated TLS libraries, including email clients.
914 PP/24 Fix segfault from trying to fprintf() to a NULL stdio FILE* if
915 using a MIME ACL for non-SMTP local injection.
916 Report and assistance in diagnosis by Warren Baker.
918 TL/08 Adjust exiqgrep to be case-insensitive for sender/receiver.
920 JH/16 Fix comparisons for 64b. Bugzilla 1385.
922 TL/09 Add expansion variable $authenticated_fail_id to keep track of
923 last id that failed so it may be referenced in subsequent ACL's.
925 TL/10 Bugzilla 1375 - Prevent TLS rebinding in ldap. Patch provided by
928 TL/11 Bugzilla 1382 - Option ldap_require_cert overrides start_tls
929 ldap library initialization, allowing self-signed CA's to be
930 used. Also properly sets require_cert option later in code by
931 using NULL (global ldap config) instead of ldap handle (per
932 session). Bug diagnosis and testing by alxgomz.
934 TL/12 Enhanced documentation in the ratelimit.pl script provided in
935 the src/util/ subdirectory.
937 TL/13 Bug 1031 - Imported transport SQL logging patch from Axel Rau
938 renamed to Transport Post Delivery Action by Jeremy Harris, as
941 TL/14 Bugzilla 1217 - Redis lookup support has been added. It is only enabled
942 when Exim is compiled with EXPERIMENTAL_REDIS. A new config variable
943 redis_servers = needs to be configured which will be used by the redis
944 lookup. Patch from Warren Baker, of The Packet Hub.
946 TL/15 Fix exiqsumm summary for corner case. Patch provided by Richard Hall.
948 TL/16 Bugzilla 1289 - Clarify host/ip processing when have errors looking up a
949 hostname or reverse DNS when processing a host list. Used suggestions
950 from multiple comments on this bug.
952 TL/17 Bugzilla 1057 - Multiple clamd TCP targets patch from Mark Zealey.
954 TL/18 Had previously added a -CONTINUE option to runtest in the test suite.
955 Missed a few lines, added it to make the runtest require no keyboard
958 TL/19 Bugzilla 1402 - Test 533 fails if any part of the path to the test suite
959 contains upper case chars. Make router use caseful_local_part.
961 TL/20 Bugzilla 1400 - Add AVOID_GNUTLS_PKCS11 build option. Allows GnuTLS
962 support when GnuTLS has been built with p11-kit.
968 PP/01 SECURITY: protect DKIM DNS decoding from remote exploit.
970 This, or similar/improved, will also be change PP/11 of 4.82.
976 PP/01 Handle short writes when writing local log-files.
977 In practice, only affects FreeBSD (8 onwards).
978 Bugzilla 1053, with thanks to Dmitry Isaikin.
980 NM/01 Bugzilla 949 - Documentation tweak
982 NM/02 Bugzilla 1093 - eximstats DATA reject detection regexps
985 NM/03 Bugzilla 1169 - primary_hostname spelling was incorrect in docs.
987 PP/02 Implemented gsasl authenticator.
989 PP/03 Implemented heimdal_gssapi authenticator with "server_keytab" option.
991 PP/04 Local/Makefile support for (AUTH|LOOKUP)_*_PC=foo to use
992 `pkg-config foo` for cflags/libs.
994 PP/05 Swapped $auth1/$auth2 for gsasl GSSAPI mechanism, to be more consistent
995 with rest of GSASL and with heimdal_gssapi.
997 PP/06 Local/Makefile support for USE_(GNUTLS|OPENSSL)_PC=foo to use
998 `pkg-config foo` for cflags/libs for the TLS implementation.
1000 PP/07 New expansion variable $tls_bits; Cyrus SASL server connection
1001 properties get this fed in as external SSF. A number of robustness
1002 and debugging improvements to the cyrus_sasl authenticator.
1004 PP/08 cyrus_sasl server now expands the server_realm option.
1006 PP/09 Bugzilla 1214 - Log authentication information in reject log.
1007 Patch by Jeremy Harris.
1009 PP/10 Added dbmjz lookup type.
1011 PP/11 Let heimdal_gssapi authenticator take a SASL message without an authzid.
1013 PP/12 MAIL args handles TAB as well as SP, for better interop with
1014 non-compliant senders.
1015 Analysis and variant patch by Todd Lyons.
1017 NM/04 Bugzilla 1237 - fix cases where printf format usage not indicated
1018 Bug report from Lars Müller <lars@samba.org> (via SUSE),
1019 Patch from Dirk Mueller <dmueller@suse.com>
1021 PP/13 tls_peerdn now print-escaped for spool files.
1022 Observed some $tls_peerdn in wild which contained \n, which resulted
1023 in spool file corruption.
1025 PP/14 TLS fixes for OpenSSL: support TLS 1.1 & 1.2; new "openssl_options"
1026 values; set SSL_MODE_AUTO_RETRY so that OpenSSL will retry a read
1027 or write after TLS renegotiation, which otherwise led to messages
1030 TK/01 Bugzilla 1239 - fix DKIM verification when signature was not inserted
1031 as a tracking header (ie: a signed header comes before the signature).
1032 Patch from Wolfgang Breyha.
1034 JH/01 Bugzilla 660 - Multi-valued attributes from ldap now parseable as a
1035 comma-sep list; embedded commas doubled.
1037 JH/02 Refactored ACL "verify =" logic to table-driven dispatch.
1039 PP/15 LDAP: Check for errors of TLS initialisation, to give correct
1041 Report and patch from Dmitry Banschikov.
1043 PP/16 Removed "dont_insert_empty_fragments" fron "openssl_options".
1044 Removed SSL_clear() after SSL_new() which led to protocol negotiation
1045 failures. We appear to now support TLS1.1+ with Exim.
1047 PP/17 OpenSSL: new expansion var $tls_sni, which if used in tls_certificate
1048 lets Exim select keys and certificates based upon TLS SNI from client.
1049 Also option tls_sni on SMTP Transports. Also clear $tls_bits correctly
1050 before an outbound SMTP session. New log_selector, +tls_sni.
1052 PP/18 Bugzilla 1122 - check localhost_number expansion for failure, avoid
1053 NULL dereference. Report and patch from Alun Jones.
1055 PP/19 DNS resolver init changes for NetBSD compatibility. (Risk of breakage
1056 on less well tested platforms). Obviates NetBSD pkgsrc patch-ac.
1057 Not seeing resolver debug output on NetBSD, but suspect this is a
1058 resolver implementation change.
1060 PP/20 Revert part of NM/04, it broke log_path containing %D expansions.
1061 Left warnings. Added "eximon gdb" invocation mode.
1063 PP/21 Defaulting "accept_8bitmime" to true, not false.
1065 PP/22 Added -bw for inetd wait mode support.
1067 PP/23 Added PCRE_CONFIG=yes support to Makefile for using pcre-config to
1068 locate the relevant includes and libraries. Made this the default.
1070 PP/24 Fixed headers_only on smtp transports (was not sending trailing dot).
1071 Bugzilla 1246, report and most of solution from Tomasz Kusy.
1073 JH/03 ${eval } now uses 64-bit and supports a "g" suffix (like to "k" and "m").
1074 This may cause build issues on older platforms.
1076 PP/25 Revamped GnuTLS support, passing tls_require_ciphers to
1077 gnutls_priority_init, ignoring Exim options gnutls_require_kx,
1078 gnutls_require_mac & gnutls_require_protocols (no longer supported).
1079 Added SNI support via GnuTLS too.
1080 Made ${randint:..} supplier available, if using not-too-old GnuTLS.
1082 PP/26 Added EXPERIMENTAL_OCSP for OpenSSL.
1084 PP/27 Applied dnsdb SPF support patch from Janne Snabb.
1085 Applied second patch from Janne, implementing suggestion to default
1086 multiple-strings-in-record handling to match SPF spec.
1088 JH/04 Added expansion variable $tod_epoch_l for a higher-precision time.
1090 PP/28 Fix DCC dcc_header content corruption (stack memory referenced,
1091 read-only, out of scope).
1092 Patch from Wolfgang Breyha, report from Stuart Northfield.
1094 PP/29 Fix three issues highlighted by clang analyser static analysis.
1095 Only crash-plausible issue would require the Cambridge-specific
1096 iplookup router and a misconfiguration.
1097 Report from Marcin Mirosław.
1099 PP/30 Another attempt to deal with PCRE_PRERELEASE, this one less buggy.
1101 PP/31 %D in printf continues to cause issues (-Wformat=security), so for
1102 now guard some of the printf checks behind WANT_DEEPER_PRINTF_CHECKS.
1103 As part of this, removing so much warning spew let me fix some minor
1104 real issues in debug logging.
1106 PP/32 GnuTLS was always using default tls_require_ciphers, due to a missing
1107 assignment on my part. Fixed.
1109 PP/33 Added tls_dh_max_bits option, defaulting to current hard-coded limit
1110 of NSS, for GnuTLS/NSS interop. Problem root cause diagnosis by
1111 Janne Snabb (who went above and beyond: thank you).
1113 PP/34 Validate tls_require_ciphers on startup, since debugging an invalid
1114 string otherwise requires a connection and a bunch more work and it's
1115 relatively easy to get wrong. Should also expose TLS library linkage
1118 PP/35 Pull in <features.h> on Linux, for some portability edge-cases of
1119 64-bit ${eval} (JH/03).
1121 PP/36 Define _GNU_SOURCE in exim.h; it's needed for some releases of
1122 GNU libc to support some of the 64-bit stuff, should not lead to
1123 conflicts. Defined before os.h is pulled in, so if a given platform
1124 needs to override this, it can.
1126 PP/37 Unbreak Cyrus SASL auth: SSF retrieval was incorrect, Exim thought
1127 protection layer was required, which is not implemented.
1128 Bugzilla 1254, patch from Wolfgang Breyha.
1130 PP/38 Overhaul DH prime handling, supply RFC-specified DH primes as built
1131 into Exim, default to IKE id 23 from RFC 5114 (2048 bit). Make
1132 tls_dhparam take prime identifiers. Also unbreak combination of
1133 OpenSSL+DH_params+TLSSNI.
1135 PP/39 Disable SSLv2 by default in OpenSSL support.
1141 PP/01 Solaris build fix for Oracle's LDAP libraries.
1142 Bugzilla 1109, patch from Stephen Usher.
1144 TF/01 HP/UX build fix: avoid arithmetic on a void pointer.
1146 TK/01 DKIM Verification: Fix relaxed canon for empty headers w/o
1149 TF/02 Fix a couple more cases where we did not log the error message
1150 when unlink() failed. See also change 4.74-TF/03.
1152 TF/03 Make the exiwhat support code safe for signals. Previously Exim might
1153 lock up or crash if it happened to be inside a call to libc when it
1154 got a SIGUSR1 from exiwhat.
1156 The SIGUSR1 handler appends the current process status to the process
1157 log which is later printed by exiwhat. It used to use the general
1158 purpose logging code to do this, but several functions it calls are
1159 not safe for signals.
1161 The new output code in the SIGUSR1 handler is specific to the process
1162 log, and simple enough that it's easy to inspect for signal safety.
1163 Removing some special cases also simplifies the general logging code.
1164 Removing the spurious timestamps from the process log simplifies
1167 TF/04 Improved ratelimit ACL condition.
1169 The /noupdate option has been deprecated in favour of /readonly which
1170 has clearer semantics. The /leaky, /strict, and /readonly update modes
1171 are mutually exclusive. The update mode is no longer included in the
1172 database key; it just determines when the database is updated. (This
1173 means that when you upgrde Exim will forget old rate measurements.)
1175 Exim now checks that the per_* options are used with an update mode that
1176 makes sense for the current ACL. For example, when Exim is processing a
1177 message (e.g. acl_smtp_rcpt or acl_smtp_data, etc.) you can specify
1178 per_mail/leaky or per_mail/strict; otherwise (e.g. in acl_smtp_helo) you
1179 must specify per_mail/readonly. If you omit the update mode it defaults to
1180 /leaky where that makes sense (as before) or /readonly where required.
1182 The /noupdate option is now undocumented but still supported for
1183 backwards compatibility. It is equivalent to /readonly except that in
1184 ACLs where /readonly is required you may specify /leaky/noupdate or
1185 /strict/noupdate which are treated the same as /readonly.
1187 A useful new feature is the /count= option. This is a generalization
1188 of the per_byte option, so that you can measure the throughput of other
1189 aggregate values. For example, the per_byte option is now equivalent
1190 to per_mail/count=${if >{0}{$message_size} {0} {$message_size} }.
1192 The per_rcpt option has been generalized using the /count= mechanism
1193 (though it's more complicated than the per_byte equivalence). When it is
1194 used in acl_smtp_rcpt, the per_rcpt option adds recipients to the
1195 measured rate one at a time; if it is used later (e.g. in acl_smtp_data)
1196 or in a non-SMTP ACL it adds all the recipients in one go. (The latter
1197 /count=$recipients_count behaviour used to work only in non-SMTP ACLs.)
1198 Note that using per_rcpt with a non-readonly update mode in more than
1199 one ACL will cause the recipients to be double-counted. (The per_mail
1200 and per_byte options don't have this problem.)
1202 The handling of very low rates has changed slightly. If the computed rate
1203 is less than the event's count (usually one) then this event is the first
1204 after a long gap. In this case the rate is set to the same as this event's
1205 count, so that the first message of a spam run is counted properly.
1207 The major new feature is a mechanism for counting the rate of unique
1208 events. The new per_addr option counts the number of different
1209 recipients that someone has sent messages to in the last time period. It
1210 behaves like per_rcpt if all the recipient addresses are different, but
1211 duplicate recipient addresses do not increase the measured rate. Like
1212 the /count= option this is a general mechanism, so the per_addr option
1213 is equivalent to per_rcpt/unique=$local_part@$domain. You can, for
1214 example, measure the rate that a client uses different sender addresses
1215 with the options per_mail/unique=$sender_address. There are further
1216 details in the main documentation.
1218 TF/05 Removed obsolete $Cambridge$ CVS revision strings.
1220 TF/06 Removed a few PCRE remnants.
1222 TF/07 Automatically extract Exim's version number from tags in the git
1223 repository when doing development or release builds.
1225 PP/02 Raise smtp_cmd_buffer_size to 16kB.
1226 Bugzilla 879. Patch from Paul Fisher.
1228 PP/03 Implement SSL-on-connect outbound with protocol=smtps on smtp transport.
1229 Heavily based on revision 40f9a89a from Simon Arlott's tree.
1232 PP/04 Use .dylib instead of .so for dynamic library loading on MacOS.
1234 PP/05 Variable $av_failed, true if the AV scanner deferred.
1235 Bugzilla 1078. Patch from John Horne.
1237 PP/06 Stop make process more reliably on build failure.
1238 Bugzilla 1087. Patch from Heiko Schlittermann.
1240 PP/07 Make maildir_use_size_file an _expandable_ boolean.
1241 Bugzilla 1089. Patch from Heiko Schlittermann.
1243 PP/08 Handle ${run} returning more data than OS pipe buffer size.
1244 Bugzilla 1131. Patch from Holger Weiß.
1246 PP/09 Handle IPv6 addresses with SPF.
1247 Bugzilla 860. Patch from Wolfgang Breyha.
1249 PP/10 GnuTLS: support TLS 1.2 & 1.1.
1251 Use gnutls_certificate_verify_peers2() [patch from Andreas Metzler].
1254 PP/11 match_* no longer expand right-hand-side by default.
1255 New compile-time build option, EXPAND_LISTMATCH_RHS.
1256 New expansion conditions, "inlist", "inlisti".
1258 PP/12 fix uninitialised greeting string from PP/03 (smtps client support).
1260 PP/13 shell and compiler warnings fixes for RC1-RC4 changes.
1262 PP/14 fix log_write() format string regression from TF/03.
1263 Bugzilla 1152. Patch from Dmitry Isaikin.
1269 PP/01 The new ldap_require_cert option would segfault if used. Fixed.
1271 PP/02 Harmonised TLS library version reporting; only show if debugging.
1272 Layout now matches that introduced for other libraries in 4.74 PP/03.
1274 PP/03 New openssl_options items: no_sslv2 no_sslv3 no_ticket no_tlsv1
1276 PP/04 New "dns_use_edns0" global option.
1278 PP/05 Don't segfault on misconfiguration of ref:name exim-user as uid.
1281 PP/06 Extra paranoia around buffer usage at the STARTTLS transition.
1282 nb: Exim is not vulnerable to http://www.kb.cert.org/vuls/id/555316
1284 TK/01 Updated PolarSSL code to 0.14.2.
1285 Bugzilla 1097. Patch from Andreas Metzler.
1287 PP/07 Catch divide-by-zero in ${eval:...}.
1288 Fixes bugzilla 1102.
1290 PP/08 Condition negation of bool{}/bool_lax{} did not negate. Fixed.
1293 TK/02 Bugzilla 1106: CVE-2011-1764 - DKIM log line was subject to a
1294 format-string attack -- SECURITY: remote arbitrary code execution.
1296 TK/03 SECURITY - DKIM signature header parsing was double-expanded, second
1297 time unintentionally subject to list matching rules, letting the header
1298 cause arbitrary Exim lookups (of items which can occur in lists, *not*
1299 arbitrary string expansion). This allowed for information disclosure.
1301 PP/09 Fix another SIGFPE (x86) in ${eval:...} expansion, this time related to
1302 INT_MIN/-1 -- value coerced to INT_MAX.
1308 NM/01 Workround for PCRE version dependency in version reporting
1311 TF/01 Update valgrind.h and memcheck.h to copies from valgrind-3.6.0.
1312 This fixes portability to compilers other than gcc, notably
1313 Solaris CC and HP-UX CC. Fixes Bugzilla 1050.
1315 TF/02 Bugzilla 139: Avoid using the += operator in the modular lookup
1316 makefiles for portability to HP-UX and POSIX correctness.
1318 PP/01 Permit LOOKUP_foo enabling on the make command-line.
1319 Also via indented variable definition in the Makefile.
1320 (Debugging by Oliver Heesakkers).
1322 PP/02 Restore caching of spamd results with expanded spamd_address.
1323 Patch from author of expandable spamd_address patch, Wolfgang Breyha.
1325 PP/03 Build issue: lookups-Makefile now exports LC_ALL=C
1326 Improves build reliability. Fix from: Frank Elsner
1328 NM/02 Fix wide character breakage in the rfc2047 coding
1329 Fixes bug 1064. Patch from Andrey N. Oktyabrski
1331 NM/03 Allow underscore in dnslist lookups
1332 Fixes bug 1026. Patch from Graeme Fowler
1334 PP/04 Bugzilla 230: Support TLS-enabled LDAP (in addition to ldaps).
1335 Code patches from Adam Ciarcinski of NetBSD.
1337 NM/04 Fixed exiqgrep to cope with mailq missing size issue
1340 PP/05 Bugzilla 1083: when lookup expansion defers, escape the output which
1341 is logged, to avoid truncation. Patch from John Horne.
1343 PP/06 Bugzilla 1042: implement freeze_signal on pipe transports.
1344 Patch from Jakob Hirsch.
1346 PP/07 Bugzilla 1061: restrict error messages sent over SMTP to not reveal
1347 SQL string expansion failure details.
1348 Patch from Andrey Oktyabrski.
1350 PP/08 Bugzilla 486: implement %M datestamping in log filenames.
1351 Patch from Simon Arlott.
1353 PP/09 New lookups functionality failed to compile on old gcc which rejects
1354 extern declarations in function scope.
1355 Patch from Oliver Fleischmann
1357 PP/10 Use sig_atomic_t for flags set from signal handlers.
1358 Check getgroups() return and improve debugging.
1359 Fixed developed for diagnosis in bug 927 (which turned out to be
1362 PP/11 Bugzilla 1055: Update $message_linecount for maildir_tag.
1363 Patch from Mark Zealey.
1365 PP/12 Bugzilla 1056: Improved spamd server selection.
1366 Patch from Mark Zealey.
1368 PP/13 Bugzilla 1086: Deal with maildir quota file races.
1369 Based on patch from Heiko Schlittermann.
1371 PP/14 Bugzilla 1019: DKIM multiple signature generation fix.
1372 Patch from Uwe Doering, sign-off by Michael Haardt.
1374 NM/05 Fix to spam.c to accommodate older gcc versions which dislike
1375 variable declaration deep within a block. Bug and patch from
1378 PP/15 lookups-Makefile IRIX compatibilty coercion.
1380 PP/16 Make DISABLE_DKIM build knob functional.
1382 NM/06 Bugzilla 968: child_open_uid: restore default SIGPIPE handler
1383 Patch by Simon Arlott
1385 TF/03 Fix valgrind.h portability to C89 compilers that do not support
1386 variable argument macros. Our copy now differs from upstream.
1392 TF/01 Failure to get a lock on a hints database can have serious
1393 consequences so log it to the panic log.
1395 TF/02 Log LMTP confirmation messages in the same way as SMTP,
1396 controlled using the smtp_confirmation log selector.
1398 TF/03 Include the error message when we fail to unlink a spool file.
1400 DW/01 Bugzilla 139: Support dynamically loaded lookups as modules.
1401 With thanks to Steve Haslam, Johannes Berg & Serge Demonchaux
1402 for maintaining out-of-tree patches for some time.
1404 PP/01 Bugzilla 139: Documentation and portability issues.
1405 Avoid GNU Makefile-isms, let Exim continue to build on BSD.
1406 Handle per-OS dynamic-module compilation flags.
1408 PP/02 Let /dev/null have normal permissions.
1409 The 4.73 fixes were a little too stringent and complained about the
1410 permissions on /dev/null. Exempt it from some checks.
1411 Reported by Andreas M. Kirchwitz.
1413 PP/03 Report version information for many libraries, including
1414 Exim version information for dynamically loaded libraries. Created
1415 version.h, now support a version extension string for distributors
1416 who patch heavily. Dynamic module ABI change.
1418 PP/04 CVE-2011-0017 - check return value of setuid/setgid. This is a
1419 privilege escalation vulnerability whereby the Exim run-time user
1420 can cause root to append content of the attacker's choosing to
1423 PP/05 Bugzilla 1041: merged DCC maintainer's fixes for return code.
1426 PP/06 Bugzilla 1071: fix delivery logging with untrusted macros.
1427 If dropping privileges for untrusted macros, we disabled normal logging
1428 on the basis that it would fail; for the Exim run-time user, this is not
1429 the case, and it resulted in successful deliveries going unlogged.
1430 Fixed. Reported by Andreas Metzler.
1436 PP/01 Date: & Message-Id: revert to normally being appended to a message,
1437 only prepend for the Resent-* case. Fixes regression introduced in
1438 Exim 4.70 by NM/22 for Bugzilla 607.
1440 PP/02 Include check_rfc2047_length in configure.default because we're seeing
1441 increasing numbers of administrators be bitten by this.
1443 JJ/01 Added DISABLE_DKIM and comment to src/EDITME
1445 PP/03 Bugzilla 994: added openssl_options main configuration option.
1447 PP/04 Bugzilla 995: provide better SSL diagnostics on failed reads.
1449 PP/05 Bugzilla 834: provide a permit_coredump option for pipe transports.
1451 PP/06 Adjust NTLM authentication to handle SASL Initial Response.
1453 PP/07 If TLS negotiated an anonymous cipher, we could end up with SSL but
1454 without a peer certificate, leading to a segfault because of an
1455 assumption that peers always have certificates. Be a little more
1456 paranoid. Problem reported by Martin Tscholak.
1458 PP/08 Bugzilla 926: switch ClamAV to use the new zINSTREAM API for content
1459 filtering; old API available if built with WITH_OLD_CLAMAV_STREAM=yes
1460 NB: ClamAV planning to remove STREAM in "middle of 2010".
1461 CL also introduces -bmalware, various -d+acl logging additions and
1462 more caution in buffer sizes.
1464 PP/09 Implemented reverse_ip expansion operator.
1466 PP/10 Bugzilla 937: provide a "debug" ACL control.
1468 PP/11 Bugzilla 922: Documentation dusting, patch provided by John Horne.
1470 PP/12 Bugzilla 973: Implement --version.
1472 PP/13 Bugzilla 752: Refuse to build/run if Exim user is root/0.
1474 PP/14 Build without WITH_CONTENT_SCAN. Path from Andreas Metzler.
1476 PP/15 Bugzilla 816: support multiple condition rules on Routers.
1478 PP/16 Add bool_lax{} expansion operator and use that for combining multiple
1479 condition rules, instead of bool{}. Make both bool{} and bool_lax{}
1480 ignore trailing whitespace.
1482 JJ/02 prevent non-panic DKIM error from being sent to paniclog
1484 JJ/03 added tcp_wrappers_daemon_name to allow host entries other than
1487 PP/17 Fix malware regression for cmdline scanner introduced in PP/08.
1488 Notification from Dr Andrew Aitchison.
1490 PP/18 Change ClamAV response parsing to be more robust and to handle ClamAV's
1491 ExtendedDetectionInfo response format.
1492 Notification from John Horne.
1494 PP/19 OpenSSL 1.0.0a compatibility const-ness change, should be backwards
1497 PP/20 Added a CONTRIBUTING file. Fixed the documentation build to use http:
1498 XSL and documented dependency on system catalogs, with examples of how
1501 DW/21 Added Valgrind hooks in store.c to help it capture out-of-bounds store
1504 DW/22 Bugzilla 1044: CVE-2010-4345 - partial fix: restrict default behaviour
1505 of CONFIGURE_OWNER and CONFIGURE_GROUP options to no longer allow a
1506 configuration file which is writeable by the Exim user or group.
1508 DW/23 Bugzilla 1044: CVE-2010-4345 - part two: extend checks for writeability
1509 of configuration files to cover files specified with the -C option if
1510 they are going to be used with root privileges, not just the default
1513 DW/24 Bugzilla 1044: CVE-2010-4345 - part three: remove ALT_CONFIG_ROOT_ONLY
1514 option (effectively making it always true).
1516 DW/25 Add TRUSTED_CONFIG_PREFIX_FILE option to allow alternative configuration
1517 files to be used while preserving root privileges.
1519 DW/26 Set FD_CLOEXEC on SMTP sockets after forking in the daemon, to ensure
1520 that rogue child processes cannot use them.
1522 PP/27 Bugzilla 1047: change the default for system_filter_user to be the Exim
1523 run-time user, instead of root.
1525 PP/28 Add WHITELIST_D_MACROS option to let some macros be overridden by the
1526 Exim run-time user without dropping privileges.
1528 DW/29 Remove use of va_copy() which breaks pre-C99 systems. Duplicate the
1529 result string, instead of calling string_vformat() twice with the same
1532 DW/30 Allow TRUSTED_CONFIG_PREFIX_FILE only for Exim or CONFIGURE_OWNER, not
1533 for other users. Others should always drop root privileges if they use
1534 -C on the command line, even for a whitelisted configure file.
1536 DW/31 Turn TRUSTED_CONFIG_PREFIX_FILE into TRUSTED_CONFIG_FILE. No prefixes.
1538 NM/01 Fixed bug #1002 - Message loss when using multiple deliveries
1544 JJ/01 installed exipick 20100104.1, adding $max_received_linelength,
1545 $data_path, and $header_path variables; fixed documentation bugs and
1548 JJ/02 installed exipick 20100222.0, added --input-dir and --finput to allow
1549 exipick to access non-standard spools, including the "frozen" queue
1552 NM/01 Bugzilla 965: Support mysql stored procedures.
1553 Patch from Alain Williams
1555 NM/02 Bugzilla 961: Spacing fix (syntax error) on Makefile directives for NetBSD
1557 NM/03 Bugzilla 955: Documentation fix for max_rcpts.
1558 Patch from Andreas Metzler
1560 NM/04 Bugzilla 954: Fix for unknown responses from Dovecot authenticator.
1561 Patch from Kirill Miazine
1563 NM/05 Bugzilla 671: Added umask to procmail example.
1565 JJ/03 installed exipick 20100323.0, fixing doc bug
1567 NM/06 Bugzilla 988: CVE-2010-2023 - prevent hardlink attack on sticky mail
1568 directory. Notification and patch from Dan Rosenberg.
1570 TK/01 PDKIM: Upgrade PolarSSL files to upstream version 0.12.1.
1572 TK/02 Improve log output when DKIM signing operation fails.
1574 MH/01 Treat the transport option dkim_domain as a colon separated
1575 list, not as a single string, and sign the message with each element,
1576 omitting multiple occurences of the same signer.
1578 NM/07 Null terminate DKIM strings, Null initialise DKIM variable
1579 Bugzilla 985, 986. Patch by Simon Arlott
1581 NM/08 Bugzilla 967. dnsdb DNS TXT record bug fix (DKIM-related)
1582 Patch by Simon Arlott
1584 PP/01 Bugzilla 989: CVE-2010-2024 - work round race condition on
1585 MBX locking. Notification from Dan Rosenberg.
1591 TK/01 Bugzilla 912: Fix DKIM segfault on empty headers/body.
1593 NM/01 Bugzilla 913: Documentation fix for gnutls_* options.
1595 NM/02 Bugzilla 722: Documentation for randint. Better randomness defaults.
1597 NM/03 Bugzilla 847: Enable DNSDB lookup by default.
1599 NM/04 Bugzilla 915: Flag broken perl installation during build.
1605 TK/01 Added patch by Johannes Berg that expands the main option
1606 "spamd_address" if it starts with a dollar sign.
1608 TK/02 Write list of recipients to X-Envelope-Sender header when building
1609 the mbox-format spool file for content scanning (suggested by Jakob
1612 TK/03 Added patch by Wolfgang Breyha that adds experimental DCC
1613 (http://www.dcc-servers.net/) support via dccifd. Activated by
1614 setting EXPERIMENTAL_DCC=yes in Local/Makefile.
1616 TK/04 Bugzilla 673: Add f-protd malware scanner support. Patch submitted
1617 by Mark Daniel Reidel <mr@df.eu>.
1619 NM/01 Bugzilla 657: Embedded PCRE removed from the exim source tree.
1620 When building exim an external PCRE library is now needed -
1621 PCRE is a system library on the majority of modern systems.
1622 See entry on PCRE_LIBS in EDITME file.
1624 NM/02 Bugzilla 646: Removed unwanted C/R in Dovecot authenticator
1625 conversation. Added nologin parameter to request.
1626 Patch contributed by Kirill Miazine.
1628 TF/01 Do not log submission mode rewrites if they do not change the address.
1630 TF/02 Bugzilla 662: Fix stack corruption before exec() in daemon.c.
1632 NM/03 Bugzilla 602: exicyclog now handles panic log, and creates empty
1633 log files in place. Contributed by Roberto Lima.
1635 NM/04 Bugzilla 667: Close socket used by dovecot authenticator.
1637 TF/03 Bugzilla 615: When checking the local_parts router precondition
1638 after a local_part_suffix or local_part_prefix option, Exim now
1639 does not use the address's named list lookup cache, since this
1640 contains cached lookups for the whole local part.
1642 NM/05 Bugzilla 521: Integrated SPF Best Guess support contributed by
1643 Robert Millan. Documentation is in experimental-spec.txt.
1645 TF/04 Bugzilla 668: Fix parallel build (make -j).
1647 NM/05.2 Bugzilla 437: Prevent Maildir aux files being created with mode 000.
1649 NM/05.3 Bugzilla 598: Improvement to Dovecot authenticator handling.
1650 Patch provided by Jan Srzednicki.
1652 TF/05 Leading white space used to be stripped from $spam_report which
1653 wrecked the formatting. Now it is preserved.
1655 TF/06 Save $spam_score, $spam_bar, and $spam_report in spool files, so
1656 that they are available at delivery time.
1658 TF/07 Fix the way ${extract is skipped in the untaken branch of a conditional.
1660 TF/08 TLS error reporting now respects the incoming_interface and
1661 incoming_port log selectors.
1663 TF/09 Produce a more useful error message if an SMTP transport's hosts
1664 setting expands to an empty string.
1666 NM/06 Bugzilla 744: EXPN did not work under TLS.
1667 Patch provided by Phil Pennock.
1669 NM/07 Bugzilla 769: Extraneous comma in usage fprintf
1670 Patch provided by Richard Godbee.
1672 NM/08 Fixed erroneous documentation references to smtp_notquit_acl to be
1673 acl_smtp_notquit, added index entry.
1675 NM/09 Bugzilla 787: Potential buffer overflow in string_format.
1676 Patch provided by Eugene Bujak.
1678 NM/10 Bugzilla 770: Problem on some platforms modifying the len parameter to
1679 accept(). Patch provided by Maxim Dounin.
1681 NM/11 Bugzilla 749: Preserve old behaviour of blanks comparing equal to zero.
1682 Patch provided by Phil Pennock.
1684 NM/12 Bugzilla 497: Correct behaviour of exiwhat when no config exists.
1686 NM/13 Bugzilla 590: Correct handling of Resent-Date headers.
1687 Patch provided by Brad "anomie" Jorsch.
1689 NM/14 Bugzilla 622: Added timeout setting to transport filter.
1690 Patch provided by Dean Brooks.
1692 TK/05 Add native DKIM support (does not depend on external libraries).
1694 NM/15 Bugzilla 854: Removed code that symlinks to pcre as its no longer useful.
1695 Patch provided by Graeme Fowler.
1697 NM/16 Bugzilla 851: Documentation example syntax fix.
1699 NM/17 Changed NOTICE file to remove references to embedded PCRE.
1701 NM/18 Bugzilla 894: Fix issue with very long lines including comments in
1704 NM/19 Bugzilla 745: TLS version reporting.
1705 Patch provided by Phil Pennock.
1707 NM/20 Bugzilla 167: bool: condition support.
1708 Patch provided by Phil Pennock.
1710 NM/21 Bugzilla 665: gnutls_compat_mode to allow compatibility with broken
1711 clients. Patch provided by Phil Pennock.
1713 NM/22 Bugzilla 607: prepend (not append) Resent-Message-ID and Resent-Date.
1714 Patch provided by Brad "anomie" Jorsch.
1716 NM/23 Bugzilla 687: Fix misparses in eximstats.
1717 Patch provided by Heiko Schlittermann.
1719 NM/24 Bugzilla 688: Fix exiwhat to handle log_selector = +pid.
1720 Patch provided by Heiko Schlittermann.
1722 NM/25 Bugzilla 727: Use transport mode as default mode for maildirsize file.
1723 plus update to original patch.
1725 NM/26 Bugzilla 799: Documentation correction for ratelimit.
1727 NM/27 Bugzilla 802: Improvements to local interface IP addr detection.
1728 Patch provided by David Brownlee.
1730 NM/28 Bugzilla 807: Improvements to LMTP delivery logging.
1732 NM/29 Bugzilla 862, 866, 875: Documentation bugfixes.
1734 NM/30 Bugzilla 888: TLS documentation bugfixes.
1736 NM/31 Bugzilla 896: Dovecot buffer overrun fix.
1738 NM/32 Bugzilla 889: Change all instances of "expr" in shell scripts to "expr --"
1739 Unlike the original bugzilla I have changed all shell scripts in src tree.
1741 NM/33 Bugzilla 898: Transport filter timeout fix.
1742 Patch by Todd Rinaldo.
1744 NM/34 Bugzilla 901: Fix sign/unsigned and UTF mismatches.
1745 Patch by Serge Demonchaux.
1747 NM/35 Bugzilla 39: Base64 decode bug fixes.
1748 Patch by Jakob Hirsch.
1750 NM/36 Bugzilla 909: Correct connect() call in dcc code.
1752 NM/37 Bugzilla 910: Correct issue with relaxed/simple handling.
1754 NM/38 Bugzilla 908: Removed NetBSD3 support as no longer needed.
1756 NM/39 Bugzilla 911: Fixed MakeLinks build script.
1762 TK/01 Add preliminary DKIM support. Currently requires a forked version of
1763 ALT-N's libdkim that I have put here:
1764 http://duncanthrax.net/exim-experimental/
1766 Note to Michael Haardt: I had to rename some vars in sieve.c. They
1767 were called 'true' and it seems that C99 defines that as a reserved
1768 keyword to be used with 'bool' variable types. That means you could
1769 not include C99-style headers which use bools without triggering
1770 build errors in sieve.c.
1772 NM/01 Bugzilla 592: --help option is handled incorrectly if exim is invoked
1773 as mailq or other aliases. Changed the --help handling significantly
1774 to do whats expected. exim_usage() emits usage/help information.
1776 SC/01 Added the -bylocaldomain option to eximstats.
1778 NM/02 Bugzilla 619: Defended against bad data coming back from gethostbyaddr.
1780 NM/03 Bugzilla 613: Documentation fix for acl_not_smtp.
1782 NM/04 Bugzilla 628: PCRE update to 7.4 (work done by John Hall).
1788 PH/01 Another patch from the Sieve maintainer.
1790 PH/02 When an IPv6 address is converted to a string for single-key lookup
1791 in an address list (e.g. for an item such as "net24-dbm;/net/works"),
1792 dots are used instead of colons so that keys in lsearch files need not
1793 contain colons. This was done some time before quoting was made available
1794 in lsearch files. However, iplsearch files do require colons in IPv6 keys
1795 (notated using the quote facility) so as to distinguish them from IPv4
1796 keys. This meant that lookups for IP addresses in host lists did not work
1797 for iplsearch lookups.
1799 This has been fixed by arranging for IPv6 addresses to be expressed with
1800 colons if the lookup type is iplsearch. This is not incompatible, because
1801 previously such lookups could never work.
1803 The situation is now rather anomolous, since one *can* have colons in
1804 ordinary lsearch keys. However, making the change in all cases is
1805 incompatible and would probably break a number of configurations.
1807 TK/01 Change PRVS address formatting scheme to reflect latests BATV draft
1810 MH/01 The "spam" ACL condition code contained a sscanf() call with a %s
1811 conversion specification without a maximum field width, thereby enabling
1812 a rogue spamd server to cause a buffer overflow. While nobody in their
1813 right mind would setup Exim to query an untrusted spamd server, an
1814 attacker that gains access to a server running spamd could potentially
1815 exploit this vulnerability to run arbitrary code as the Exim user.
1817 TK/02 Bugzilla 502: Apply patch to make the SPF-Received: header use
1818 $primary_hostname instead of what libspf2 thinks the hosts name is.
1820 MH/02 The dsearch lookup now uses lstat(2) instead of stat(2) to look for
1821 a directory entry by the name of the lookup key. Previously, if a
1822 symlink pointed to a non-existing file or a file in a directory that
1823 Exim lacked permissions to read, a lookup for a key matching that
1824 symlink would fail. Now it is enough that a matching directory entry
1825 exists, symlink or not. (Bugzilla 503.)
1827 PH/03 The body_linecount and body_zerocount variables are now exported in the
1830 PH/04 Added the $dnslist_matched variable.
1832 PH/05 Unset $tls_cipher and $tls_peerdn before making a connection as a client.
1833 This means they are set thereafter only if the connection becomes
1836 PH/06 Added the client_condition to authenticators so that some can be skipped
1837 by clients under certain conditions.
1839 PH/07 The error message for a badly-placed control=no_multiline_responses left
1840 "_responses" off the end of the name.
1842 PH/08 Added -Mvc to output a copy of a message in RFC 2822 format.
1844 PH/09 Tidied the code for creating ratelimiting keys, creating them explicitly
1845 (without spaces) instead of just copying the configuration text.
1847 PH/10 Added the /noupdate option to the ratelimit ACL condition.
1849 PH/11 Added $max_received_linelength.
1851 PH/12 Added +ignore_defer and +include_defer to host lists.
1853 PH/13 Installed PCRE version 7.2. This needed some changes because of the new
1854 way in which PCRE > 7.0 is built.
1856 PH/14 Implemented queue_only_load_latch.
1858 PH/15 Removed an incorrect (int) cast when reading the value of SIZE in a
1859 MAIL command. The effect was to mangle the value on 64-bit systems.
1861 PH/16 Another patch from the Sieve maintainer.
1863 PH/17 Added the NOTQUIT ACL, based on a patch from Ted Cooper.
1865 PH/18 If a system quota error occurred while trying to create the file for
1866 a maildir delivery, the message "Mailbox is full" was not appended to the
1867 bounce if the delivery eventually timed out. Change 4.67/27 below applied
1868 only to a quota excession during the actual writing of the file.
1870 PH/19 It seems that peer DN values may contain newlines (and other non-printing
1871 characters?) which causes problems in log lines. The DN values are now
1872 passed through string_printing() before being added to log lines.
1874 PH/20 Added the "servers=" facility to MySQL and PostgreSQL lookups. (Oracle
1875 and InterBase are left for another time.)
1877 PH/21 Added message_body_newlines option.
1879 PH/22 Guard against possible overflow in moan_check_errorcopy().
1881 PH/23 POSIX allows open() to be a macro; guard against that.
1883 PH/24 If the recipient of an error message contained an @ in the local part
1884 (suitably quoted, of course), incorrect values were put in $domain and
1885 $local_part during the evaluation of errors_copy.
1891 MH/01 Fix for bug #448, segfault in Dovecot authenticator when interface_address
1892 is unset (happens when testing with -bh and -oMi isn't used). Thanks to
1895 PH/01 Added a new log selector smtp_no_mail, to log SMTP sessions that do not
1896 issue a MAIL command.
1898 PH/02 In an ACL statement such as
1900 deny dnslists = X!=127.0.0.2 : X=127.0.0.2
1902 if a client was not listed at all, or was listed with a value other than
1903 127.0.0.2, in the X list, but was listed with 127.0.0.2 in the Y list,
1904 the condition was not true (as it should be), so access was not denied.
1905 The bug was that the ! inversion was incorrectly passed on to the second
1906 item. This has been fixed.
1908 PH/03 Added additional dnslists conditions == and =& which are different from
1909 = and & when the dns lookup returns more than one IP address.
1911 PH/04 Added gnutls_require_{kx,mac,protocols} to give more control over the
1912 cipher suites used by GnuTLS. These options are ignored by OpenSSL.
1914 PH/05 After discussion on the list, added a compile time option ENABLE_DISABLE_
1915 FSYNC, which compiles an option called disable_fsync that allows for
1916 bypassing fsync(). The documentation is heavily laced with warnings.
1918 SC/01 Updated eximstats to collate all SpamAssassin rejects into one bucket.
1920 PH/06 Some tidies to the infrastructure of the Test Suite that is concerned
1921 with the auxiliary C programs that it uses: (1) Arrange for BIND_8_COMPAT
1922 to be defined when compiling on OSX (Darwin); (2) Tidies to the Makefile,
1923 including adding "make clean"; (3) Added -fPIC when compiling the test
1924 dynamically loaded module, to get rid of a warning.
1926 MH/02 Fix for bug #451, causing paniclog entries to be written if a bounce
1927 message fails, move_frozen_messages = true and ignore_bounce_errors_after
1928 = 0s. The bug is otherwise harmless.
1930 PH/07 There was a bug in the dovecot authenticator such that the value of
1931 $auth1 could be overwritten, and so not correctly preserved, after a
1932 successful authentication. This usually meant that the value preserved by
1933 the server_setid option was incorrect.
1935 PH/08 Added $smtp_count_at_connection_start, deliberately with a long name.
1937 PH/09 Installed PCRE release 7.0.
1939 PH/10 The acl_not_smtp_start ACL was, contrary to the documentation, not being
1940 run for batched SMTP input. It is now run at the start of every message
1941 in the batch. While fixing this I discovered that the process information
1942 (output by running exiwhat) was not always getting set for -bs and -bS
1943 input. This is fixed, and it now also says "batched" for BSMTP.
1945 PH/11 Added control=no_pipelining.
1947 PH/12 Added $sending_ip_address and $sending_port (mostly Magnus Holmgren's
1948 patch, slightly modified), and move the expansion of helo_data till after
1949 the connection is made in the smtp transport (so it can use these
1952 PH/13 Added ${rfc2047d: to decoded RFC 2047 strings.
1954 PH/14 Added log_selector = +pid.
1956 PH/15 Flush SMTP output before delaying, unless control=no_delay_flush is set.
1958 PH/16 Add ${if forany and ${if forall.
1960 PH/17 Added dsn_from option to vary the From: line in DSNs.
1962 PH/18 Flush SMTP output before performing a callout, unless control =
1963 no_callout_flush is set.
1965 PH/19 Change 4.64/PH/36 introduced a bug: when address_retry_include_sender
1966 was true (the default) a successful delivery failed to delete the retry
1967 item, thus causing premature timeout of the address. The bug is now
1970 PH/20 Added hosts_avoid_pipelining to the smtp transport.
1972 PH/21 Long custom messages for fakedefer and fakereject are now split up
1973 into multiline reponses in the same way that messages for "deny" and
1974 other ACL rejections are.
1976 PH/22 Applied Jori Hamalainen's speed-up changes and typo fixes to exigrep,
1977 with slight modification.
1979 PH/23 Applied sieve patches from the maintainer "tracking the latest notify
1980 draft, changing the syntax and factoring some duplicate code".
1982 PH/24 When the log selector "outgoing_port" was set, the port was shown as -1
1983 for deliveries of the second and subsequent messages over the same SMTP
1986 PH/25 Applied Magnus Holmgren's patch for ${addresses, ${map, ${filter, and
1987 ${reduce, with only minor "tidies".
1989 SC/02 Applied Daniel Tiefnig's patch to improve the '($parent) =' pattern match.
1991 PH/26 Added a "continue" ACL modifier that does nothing, for the benefit of its
1992 expansion side effects.
1994 PH/27 When a message times out after an over-quota error from an Exim-imposed
1995 quota, the bounce message says "mailbox is full". This message was not
1996 being given when it was a system quota that was exceeded. It now should
1999 MH/03 Made $recipients available in local_scan(). local_scan() already has
2000 better access to the recipient list through recipients_list[], but
2001 $recipients can be useful in postmaster-provided expansion strings.
2003 PH/28 The $smtp_command and $smtp_command_argument variables were not correct
2004 in the case of a MAIL command with additional options following the
2005 address, for example: MAIL FROM:<foo@bar> SIZE=1234. The option settings
2006 were accidentally chopped off.
2008 PH/29 SMTP synchronization checks are implemented when a command is read -
2009 there is a check that no more input is waiting when there shouldn't be
2010 any. However, for some commands, a delay in an ACL can mean that it is
2011 some time before the response is written. In this time, more input might
2012 arrive, invalidly. So now there are extra checks after an ACL has run for
2013 HELO/EHLO and after the predata ACL, and likewise for MAIL and RCPT when
2014 pipelining has not been advertised.
2016 PH/30 MH's patch to allow iscntrl() characters to be list separators.
2018 PH/31 Unlike :fail:, a custom message specified with :defer: was not being
2019 returned in the SMTP response when smtp_return_error_details was false.
2020 This has been fixed.
2022 PH/32 Change the Dovecot authenticator to use read() and write() on the socket
2023 instead of the C I/O that was originally supplied, because problems were
2024 reported on Solaris.
2026 PH/33 Compile failed with OpenSSL 0.9.8e. This was due to a coding error in
2027 Exim which did not show up earlier: it was assuming that a call to
2028 SSL_CTX_set_info_callback() might give an error value. In fact, there is
2029 no error. In previous releases of OpenSSL, SSL_CTX_set_info_callback()
2030 was a macro that became an assignment, so it seemed to work. This has
2031 changed to a proper function call with a void return, hence the compile
2032 error. Exim's code has been fixed.
2034 PH/34 Change HDA_SIZE in oracle.c from 256 to 512. This is needed for 64-bit
2037 PH/35 Applied a patch from the Sieve maintainer which fixes a bug in "notify".
2039 PH/36 Applied John Jetmore's patch to add -v functionality to exigrep.
2041 PH/37 If a message is not accepted after it has had an id assigned (e.g.
2042 because it turns out to be too big or there is a timeout) there is no
2043 "Completed" line in the log. When some messages of this type were
2044 selected by exigrep, they were listed as "not completed". Others were
2045 picked up by some special patterns. I have improved the selection
2046 criteria to be more general.
2048 PH/38 The host_find_failed option in the manualroute router can now be set
2049 to "ignore", to completely ignore a host whose IP address cannot be
2050 found. If all hosts are ignored, the behaviour is controlled by the new
2051 host_all_ignored option.
2053 PH/39 In a list of hosts for manualroute, if one item (either because of multi-
2054 homing or because of multiple MX records with /mx) generated more than
2055 one IP address, and the following item turned out to be the local host,
2056 all the secondary addresses of the first item were incorrectly removed
2057 from the list, along with the local host and any following hosts (which
2058 is what is supposed to happen).
2060 PH/40 When Exim receives a message, it writes the login name, uid, and gid of
2061 whoever called Exim into the -H file. In the case of the daemon it was
2062 behaving confusingly. When first started, it used values for whoever
2063 started the daemon, but after a SIGHUP it used the Exim user (because it
2064 calls itself on a restart). I have changed the code so that it now always
2067 PH/41 (Following a suggestion from Tony Finch) If all the RCPT commands in a
2068 message are rejected with the same error (e.g. no authentication or bad
2069 sender address), and a DATA command is nevertheless sent (as can happen
2070 with PIPELINING or a stupid MUA), the error message that was given to the
2071 RCPT commands is included in the rejection of the DATA command. This is
2072 intended to be helpful for MUAs that show only the final error to their
2075 PH/42 Another patch from the Sieve maintainer.
2077 SC/02 Eximstats - Differentiate between permanent and temporary rejects.
2078 Eximstats - Fixed some broken HTML links and added missing column headers
2080 Eximstats - Fixed Grand Total Summary Domains, Edomains, and Email
2081 columns for Rejects, Temp Rejects, Ham, and Spam rows.
2083 SC/03 Eximstats - V1.58 Fix to get <> and blackhole to show in edomain tables.
2085 PH/43 Yet another patch from the Sieve maintainer.
2087 PH/44 I found a way to check for a TCP/IP connection going away before sending
2088 the response to the final '.' that terminates a message, but only in the
2089 case where the client has not sent further data following the '.'
2090 (unfortunately, this is allowed). However, in many cases there won't be
2091 any further data because there won't be any more messages to send. A call
2092 to select() can be used: if it shows that the input is "ready", there is
2093 either input waiting, or the socket has been closed. An attempt to read
2094 the next input character can distinguish the two cases. Previously, Exim
2095 would have sent an OK response which the client would never have see.
2096 This could lead to message repetition. This fix should cure that, at
2097 least in a lot of common cases.
2099 PH/45 Do not advertise STARTTLS in response to HELP unless it would be
2100 advertised in response to EHLO.
2106 PH/01 Two more bugs that were introduced by 4.64/PH/07, in addition to the one
2107 fixed by 4.65/MH/01 (is this a record?) are fixed:
2109 (i) An empty string was always treated as zero by the numeric comparison
2110 operators. This behaviour has been restored.
2112 (ii) It is documented that the numeric comparison operators always treat
2113 their arguments as decimal numbers. This was broken in that numbers
2114 starting with 0 were being interpreted as octal.
2116 While fixing these problems I realized that there was another issue that
2117 hadn't been noticed. Values of message_size_limit (both the global option
2118 and the transport option) were treated as octal if they started with 0.
2119 The documentation was vague. These values are now always treated as
2120 decimal, and I will make that clear in the documentation.
2126 TK/01 Disable default definition of HAVE_LINUX_SENDFILE. Clashes with
2127 Linux large file support (_FILE_OFFSET_BITS=64) on older glibc
2130 MH/01 Don't check that the operands of numeric comparison operators are
2131 integers when their expansion is in "skipping" mode (fixes bug
2132 introduced by 4.64-PH/07).
2134 PH/01 If a system filter or a router generates more than SHRT_MAX (32767)
2135 child addresses, Exim now panics and dies. Previously, because the count
2136 is held in a short int, deliveries were likely to be lost. As such a
2137 large number of recipients for a single message is ridiculous
2138 (performance will be very, very poor), I have chosen to impose a limit
2139 rather than extend the field.
2145 TK/01 Bugzilla #401. Fix DK spooling code so that it can overwrite a
2146 leftover -K file (the existence of which was triggered by #402).
2147 While we were at it, introduced process PID as part of the -K
2148 filename. This should rule out race conditions when creating
2151 TK/02 Bugzilla #402. Apply patch from Simon Arlott, speeding up DK signing
2152 processing considerably. Previous code took too long for large mails,
2153 triggering a timeout which in turn triggers #401.
2155 TK/03 Introduced HAVE_LINUX_SENDFILE to os.h-Linux. Currently only used
2156 in the DK code in transports.c. sendfile() is not really portable,
2157 hence the _LINUX specificness.
2159 TF/01 In the add_headers option to the mail command in an Exim filter,
2160 there was a bug that Exim would claim a syntax error in any
2161 header after the first one which had an odd number of characters
2164 PH/01 If a server that rejects MAIL FROM:<> was the target of a sender
2165 callout verification, Exim cached a "reject" for the entire domain. This
2166 is correct for most verifications, but it is not correct for a recipient
2167 verification with use_sender or use_postmaster set, because in that case
2168 the callout does not use MAIL FROM:<>. Exim now distinguishes the special
2169 case of MAIL FROM:<> rejection from other early rejections (e.g.
2170 rejection of HELO). When verifying a recipient using a non-null MAIL
2171 address, the cache is ignored if it shows MAIL FROM:<> rejection.
2172 Whatever the result of the callout, the value of the domain cache is
2173 left unchanged (for any other kind of callout, getting as far as trying
2174 RCPT means that the domain itself is ok).
2176 PH/02 Tidied a number of unused variable and signed/unsigned warnings that
2179 PH/03 On Solaris, an unexpectedly close socket (dropped connection) can
2180 manifest itself as EPIPE rather than ECONNECT. When tidying away a
2181 session, the daemon ignores ECONNECT errors and logs others; it now
2182 ignores EPIPE as well.
2184 PH/04 Applied Nico Erfurth's refactoring patch to tidy up mime.c
2185 (quoted-printable decoding).
2187 PH/05 Applied Nico Erfurth's refactoring patch to tidy up spool_mbox.c, and
2188 later the small subsequent patch to fix an introduced bug.
2190 PH/06 Installed the latest Cygwin Makefile from the Cygwin maintainer.
2192 PH/07 There was no check for overflow in expansions such as ${if >{1}{4096M}}.
2194 PH/08 An error is now given if message_size_limit is specified negative.
2196 PH/09 Applied and tidied up Jakob Hirsch's patch for allowing ACL variables
2197 to be given (somewhat) arbitrary names.
2199 JJ/01 exipick 20060919.0, allow for arbitrary acl_ variables introduced
2202 JJ/02 exipick 20060919.0, --show-vars args can now be regular expressions,
2203 miscellaneous code fixes
2205 PH/10 Added the log_reject_target ACL modifier to specify where to log
2208 PH/11 Callouts were setting the name used for EHLO/HELO from $smtp_active_
2209 hostname. This is wrong, because it relates to the incoming message (and
2210 probably the interface on which it is arriving) and not to the outgoing
2211 callout (which could be using a different interface). This has been
2212 changed to use the value of the helo_data option from the smtp transport
2213 instead - this is what is used when a message is actually being sent. If
2214 there is no remote transport (possible with a router that sets up host
2215 addresses), $smtp_active_hostname is used.
2217 PH/12 Installed Andrey Panin's patch to add a dovecot authenticator. Various
2218 tweaks were necessary in order to get it to work (see also 21 below):
2219 (a) The code assumed that strncpy() returns a negative number on buffer
2220 overflow, which isn't the case. Replaced with Exim's string_format()
2222 (b) There were several signed/unsigned issues. I just did the minimum
2223 hacking in of casts. There is scope for a larger refactoring.
2224 (c) The code used strcasecmp() which is not a standard C function.
2225 Replaced with Exim's strcmpic() function.
2226 (d) The code set only $1; it now sets $auth1 as well.
2227 (e) A simple test gave the error "authentication client didn't specify
2228 service in request". It would seem that Dovecot has changed its
2229 interface. Fortunately there's a specification; I followed it and
2230 changed what the client sends and it appears to be working now.
2232 PH/13 Added $message_headers_raw to provide the headers without RFC 2047
2235 PH/14 Corrected misleading output from -bv when -v was also used. Suppose the
2236 address A is aliased to B and C, where B exists and C does not. Without
2237 -v the output is "A verified" because verification stops after a
2238 successful redirection if more than one address is generated. However,
2239 with -v the child addresses are also verified. Exim was outputting "A
2240 failed to verify" and then showing the successful verification for C,
2241 with its parentage. It now outputs "B failed to verify", showing B's
2242 parentage before showing the successful verification of C.
2244 PH/15 Applied Michael Deutschmann's patch to allow DNS black list processing to
2245 look up a TXT record in a specific list after matching in a combined
2248 PH/16 It seems that the options setting for the resolver (RES_DEFNAMES and
2249 RES_DNSRCH) can affect the behaviour of gethostbyname() and friends when
2250 they consult the DNS. I had assumed they would set it the way they
2251 wanted; and indeed my experiments on Linux seem to show that in some
2252 cases they do (I could influence IPv6 lookups but not IPv4 lookups).
2253 To be on the safe side, however, I have now made the interface to
2254 host_find_byname() similar to host_find_bydns(), with an argument
2255 containing the DNS resolver options. The host_find_byname() function now
2256 sets these options at its start, just as host_find_bydns() does. The smtp
2257 transport options dns_qualify_single and dns_search_parents are passed to
2258 host_find_byname() when gethostbyname=TRUE in this transport. Other uses
2259 of host_find_byname() use the default settings of RES_DEFNAMES
2260 (qualify_single) but not RES_DNSRCH (search_parents).
2262 PH/17 Applied (a modified version of) Nico Erfurth's patch to make
2263 spool_read_header() do less string testing, by means of a preliminary
2264 switch on the second character of optional "-foo" lines. (This is
2265 overdue, caused by the large number of possibilities that now exist.
2266 Originally there were few.) While I was there, I also converted the
2267 str(n)cmp tests so they don't re-test the leading "-" and the first
2268 character, in the hope this might squeeze out yet more improvement.
2270 PH/18 Two problems with "group" syntax in header lines when verifying: (1) The
2271 flag allowing group syntax was set by the header_syntax check but not
2272 turned off, possible causing trouble later; (2) The flag was not being
2273 set at all for the header_verify test, causing "group"-style headers to
2274 be rejected. I have now set it in this case, and also caused header_
2275 verify to ignore an empty address taken from a group. While doing this, I
2276 came across some other cases where the code for allowing group syntax
2277 while scanning a header line wasn't quite right (mostly, not resetting
2278 the flag correctly in the right place). These bugs could have caused
2279 trouble for malformed header lines. I hope it is now all correct.
2281 PH/19 The functions {pwcheck,saslauthd}_verify_password() are always called
2282 with the "reply" argument non-NULL. The code, however (which originally
2283 came from elsewhere) had *some* tests for NULL when it wrote to *reply,
2284 but it didn't always do it. This confused somebody who was copying the
2285 code for some other use. I have removed all the tests.
2287 PH/20 It was discovered that the GnuTLS code had support for RSA_EXPORT, a
2288 feature that was used to support insecure browsers during the U.S. crypto
2289 embargo. It requires special client support, and Exim is probably the
2290 only MTA that supported it -- and would never use it because real RSA is
2291 always available. This code has been removed, because it had the bad
2292 effect of slowing Exim down by computing (never used) parameters for the
2293 RSA_EXPORT functionality.
2295 PH/21 On the advice of Timo Sirainen, added a check to the dovecot
2296 authenticator to fail if there's a tab character in the incoming data
2297 (there should never be unless someone is messing about, as it's supposed
2298 to be base64-encoded). Also added, on Timo's advice, the "secured" option
2299 if the connection is using TLS or if the remote IP is the same as the
2300 local IP, and the "valid-client-cert option" if a client certificate has
2303 PH/22 As suggested by Dennis Davis, added a server_condition option to *all*
2304 authenticators. This can be used for authorization after authentication
2305 succeeds. (In the case of plaintext, it servers for both authentication
2308 PH/23 Testing for tls_required and lost_connection in a retry rule didn't work
2309 if any retry times were supplied.
2311 PH/24 Exim crashed if verify=helo was activated during an incoming -bs
2312 connection, where there is no client IP address to check. In this
2313 situation, the verify now always succeeds.
2315 PH/25 Applied John Jetmore's -Mset patch.
2317 PH/26 Added -bem to be like -Mset, but loading a message from a file.
2319 PH/27 In a string expansion for a processed (not raw) header when multiple
2320 headers of the same name were present, leading whitespace was being
2321 removed from all of them, but trailing whitespace was being removed only
2322 from the last one. Now trailing whitespace is removed from each header
2323 before concatenation. Completely empty headers in a concatenation (as
2324 before) are ignored.
2326 PH/28 Fixed bug in backwards-compatibility feature of PH/09 (thanks to John
2327 Jetmore). It would have mis-read ACL variables from pre-4.61 spool files.
2329 PH/29 [Removed. This was a change that I later backed out, and forgot to
2330 correct the ChangeLog entry (that I had efficiently created) before
2331 committing the later change.]
2333 PH/30 Exim was sometimes attempting to deliver messages that had suffered
2334 address errors (4xx response to RCPT) over the same connection as other
2335 messages routed to the same hosts. Such deliveries are always "forced",
2336 so retry times are not inspected. This resulted in far too many retries
2337 for the affected addresses. The effect occurred only when there were more
2338 hosts than the hosts_max_try setting in the smtp transport when it had
2339 the 4xx errors. Those hosts that it had tried were not added to the list
2340 of hosts for which the message was waiting, so if all were tried, there
2341 was no problem. Two fixes have been applied:
2343 (i) If there are any address or message errors in an SMTP delivery, none
2344 of the hosts (tried or untried) are now added to the list of hosts
2345 for which the message is waiting, so the message should not be a
2346 candidate for sending over the same connection that was used for a
2347 successful delivery of some other message. This seems entirely
2348 reasonable: after all the message is NOT "waiting for some host".
2349 This is so "obvious" that I'm not sure why it wasn't done
2350 previously. Hope I haven't missed anything, but it can't do any
2351 harm, as the worst effect is to miss an optimization.
2353 (ii) If, despite (i), such a delivery is accidentally attempted, the
2354 routing retry time is respected, so at least it doesn't keep
2355 hammering the server.
2357 PH/31 Installed Andrew Findlay's patch to close the writing end of the socket
2358 in ${readsocket because some servers need this prod.
2360 PH/32 Added some extra debug output when updating a wait-xxx database.
2362 PH/33 The hint "could be header name not terminated by colon", which has been
2363 given for certain expansion errors for a long time, was not being given
2364 for the ${if def:h_colon_omitted{... case.
2366 PH/34 The spec says: "With one important exception, whenever a domain list is
2367 being scanned, $domain contains the subject domain." There was at least
2368 one case where this was not true.
2370 PH/35 The error "getsockname() failed: connection reset by peer" was being
2371 written to the panic log as well as the main log, but it isn't really
2372 panic-worthy as it just means the connection died rather early on. I have
2373 removed the panic log writing for the ECONNRESET error when getsockname()
2376 PH/36 After a 4xx response to a RCPT error, that address was delayed (in queue
2377 runs only) independently of the message's sender address. This meant
2378 that, if the 4xx error was in fact related to the sender, a different
2379 message to the same recipient with a different sender could confuse
2380 things. In particualar, this can happen when sending to a greylisting
2381 server, but other circumstances could also provoke similar problems.
2382 I have changed the default so that the retry time for these errors is now
2383 based a combination of the sender and recipient addresses. This change
2384 can be overridden by setting address_retry_include_sender=false in the
2387 PH/37 For LMTP over TCP/IP (the smtp transport), error responses from the
2388 remote server are returned as part of bounce messages. This was not
2389 happening for LMTP over a pipe (the lmtp transport), but now it is the
2390 same for both kinds of LMTP.
2392 PH/38 Despite being documented as not happening, Exim was rewriting addresses
2393 in header lines that were in fact CNAMEs. This is no longer the case.
2395 PH/39 If -R or -S was given with -q<time>, the effect of -R or -S was ignored,
2396 and queue runs started by the daemon processed all messages. This has
2397 been fixed so that -R and -S can now usefully be given with -q<time>.
2399 PH/40 Import PCRE release 6.7 (fixes some bugs).
2401 PH/41 Add bitwise logical operations to eval (courtesy Brad Jorsch).
2403 PH/42 Give an error if -q is specified more than once.
2405 PH/43 Renamed the variables $interface_address and $interface_port as
2406 $received_ip_address and $received_port, to make it clear that these
2407 values apply to message reception, and not to the outgoing interface when
2408 a message is delivered. (The old names remain recognized, of course.)
2410 PH/44 There was no timeout on the connect() call when using a Unix domain
2411 socket in the ${readsocket expansion. There now is.
2413 PH/45 Applied a modified version of Brad Jorsch's patch to allow "message" to
2414 be meaningful with "accept".
2416 SC/01 Eximstats V1.43
2417 Bug fix for V1.42 with -h0 specified. Spotted by Chris Lear.
2419 SC/02 Eximstats V1.44
2420 Use a glob alias rather than an array ref in the generated
2421 parser. This improves both readability and performance.
2423 SC/03 Eximstats V1.45 (Marco Gaiarin / Steve Campbell)
2424 Collect SpamAssassin and rejection statistics.
2425 Don't display local sender or destination tables unless
2426 there is data to show.
2427 Added average volumes into the top table text output.
2429 SC/04 Eximstats V1.46
2430 Collect data on the number of addresses (recipients)
2431 as well as the number of messages.
2433 SC/05 Eximstats V1.47
2434 Added 'Message too big' to the list of mail rejection
2435 reasons (thanks to Marco Gaiarin).
2437 SC/06 Eximstats V1.48
2438 Mainlog lines which have GMT offsets and are too short to
2439 have a flag are now skipped.
2441 SC/07 Eximstats V1.49 (Alain Williams)
2442 Added the -emptyok flag.
2444 SC/08 Eximstats V1.50
2445 Fixes for obtaining the IP address from reject messages.
2447 JJ/03 exipick.20061117.2, made header handling as similar to exim as possible
2448 (added [br]h_ prefixes, implemented RFC2047 decoding. Fixed
2449 whitesspace changes from 4.64-PH/27
2451 JJ/04 exipick.20061117.2, fixed format and added $message_headers_raw to
2454 JJ/05 exipick.20061117.2, bug fixes (error out sooner when invalid criteria
2455 are found, allow negative numbers in numeric criteria)
2457 JJ/06 exipick.20061117.2, added new $message_body_missing variable
2459 JJ/07 exipick.20061117.2, added $received_ip_address and $received_port
2460 to match changes made in 4.64-PH/43
2462 PH/46 Applied Jori Hamalainen's patch to add features to exiqsumm.
2464 PH/47 Put in an explicit test for a DNS lookup of an address record where the
2465 "domain" is actually an IP address, and force a failure. This locks out
2466 those revolvers/nameservers that support "A-for-A" lookups, in
2467 contravention of the specifications.
2469 PH/48 When a host name was looked up from an IP address, and the subsequent
2470 forward lookup of the name timed out, the host name was left in
2471 $sender_host_name, contrary to the specification.
2473 PH/49 Although default lookup types such as lsearch* or cdb*@ have always been
2474 restricted to single-key lookups, Exim was not diagnosing an error if
2475 * or *@ was used with a query-style lookup.
2477 PH/50 Increased the value of DH_BITS in tls-gnu.c from 768 to 1024.
2479 MH/01 local_scan ABI version incremented to 1.1. It should have been updated
2480 long ago, but noone interested enough thought of it. Let's just say that
2481 the "1.1" means that there are some new functions that weren't there at
2482 some point in the past.
2484 PH/51 Error processing for expansion failure of helo_data from an smtp
2485 transport during callout processing was broken.
2487 PH/52 Applied John Jetmore's patch to allow tls-on-connect and STARTTLS to be
2488 tested/used via the -bh/-bhc/-bs options.
2490 PH/53 Added missing "#include <time.h>" to pcre/pcretest.c (this was a PCRE
2491 bug, fixed in subsequent PCRE releases).
2493 PH/54 Applied Robert Bannocks' patch to avoid a problem with references that
2494 arises when using the Solaris LDAP libraries (but not with OpenLDAP).
2496 PH/55 Check for a ridiculously long file name in exim_dbmbuild.
2502 SC/01 Use a glob alias rather than an array ref in eximstats generated
2503 parser. This improves both readability and performance.
2505 SC/02 Collect SpamAssassin and rejection statistics in eximstats.
2506 Don't display local sender or destination tables in eximstats unless
2507 there is data to show.
2508 Added average volumes into the eximstats top table text output.
2510 SC/03 Collect data on the number of addresses (recipients) as well
2511 as the number of messages in eximstats.
2513 TF/01 Correct an error in the documentation for the redirect router. Exim
2514 does (usually) call initgroups() when daemonizing.
2516 TF/02 Call initgroups() when dropping privilege in exim.c, so that Exim runs
2517 with consistent privilege compared to when running as a daemon.
2519 TF/03 Note in the spec that $authenticated_id is not set for local
2520 submissions from trusted users.
2522 TF/04 The ratelimit per_rcpt option now works correctly in acl_not_smtp.
2523 Thanks to Dean Brooks <dean@iglou.com> for the patch.
2525 TF/05 Make it easier to get SMTP authentication and TLS/SSL support working
2526 by adding some example configuration directives to the default
2527 configuration file. A little bit of work is required to uncomment the
2528 directives and define how usernames and passwords are checked, but
2529 there is now a framework to start from.
2531 PH/01 Added #define LDAP_DEPRECATED 1 to ldap.c because some of the "old"
2532 functions that Exim currently uses aren't defined in ldap.h for OpenLDAP
2533 without this. I don't know how relevant this is to other LDAP libraries.
2535 PH/02 Add the verb name to the "unknown ACL verb" error.
2537 PH/03 Magnus Holmgren's patch for filter_prepend_home.
2539 PH/03 Fixed Bugzilla #101: macro definition between ACLs doesn't work.
2541 PH/04 Applied Magnus Holmgren's patch to fix Bugzilla #98: transport's home
2542 directory not expanded when it should be if an expanded home directory
2543 was set for the address (which is overridden by the transport).
2545 PH/05 Applied Alex Kiernan's patch to fix Bugzilla #99: a problem with
2548 PH/06 Added acl_not_smtp_start, based on Johannes Berg's patch, and set the
2549 bit to forbid control=suppress_local_fixups in the acl_not_smtp ACL,
2550 because it is too late at that time, and has no effect.
2552 PH/07 Changed ${quote_pgsql to quote ' as '' instead of \' because of a
2553 security issue with \' (bugzilla #107). I could not use the
2554 PQescapeStringConn() function, because it needs a PGconn value as one of
2557 PH/08 When testing addresses using -bt, indicate those final addresses that
2558 are duplicates that would not cause an additional delivery. At least one
2559 person was confused, thinking that -bt output corresponded to deliveries.
2560 (Suppressing duplicates isn't a good idea as you lose the information
2561 about possibly different redirections that led to the duplicates.)
2563 PH/09 Applied patch from Erik to use select() instead of poll() in spam.c on
2564 systems where poll() doesn't work, in particular OS X.
2566 PH/10 Added more information to debugging output for retry time not reached.
2568 PH/11 Applied patch from Arkadiusz Miskiewicz to apply a timeout to read
2569 operations in malware.c.
2571 PH/12 Applied patch from Magnus Holmgren to include the "h" tag in Domain Keys
2574 PH/13 If write_rejectlog was set false when logging was sent to syslog with
2575 syslog_duplication set false, log lines that would normally be written
2576 both the the main log and to the reject log were not written to syslog at
2579 PH/14 In the default configuration, change the use of "message" in ACL warn
2580 statements to "add_header".
2582 PH/15 Diagnose a filter syntax error for "seen", "unseen", or "noerror" if not
2583 not followed by a command (e.g. "seen endif").
2585 PH/16 Recognize SMTP codes at the start of "message" in ACLs and after :fail:
2586 and :defer: in a redirect router. Add forbid_smtp_code to suppress the
2589 PH/17 Added extra conditions to the default value of delay_warning_condition
2593 { !eq{$h_list-id:$h_list-post:$h_list-subscribe:}{} } \
2594 { match{$h_precedence:}{(?i)bulk|list|junk} } \
2595 { match{$h_auto-submitted:}{(?i)auto-generated|auto-replied} } \
2598 The Auto-Submitted: and various List- headers are standardised, whereas I
2599 don't think Precedence: ever was.
2601 PH/18 Refactored debugging code in route_finduser() to show more information,
2602 in particular, the error code if getpwnam() issues one.
2604 PH/19 Added PQsetClientEncoding(conn, "SQL_ASCII") to the pgsql code module.
2605 This is apparently needed in addition to the PH/07 change above to avoid
2606 any possible encoding problems.
2608 PH/20 Perl can change the locale. Exim was resetting it after a ${perl call,
2609 but not after initializing Perl.
2611 PH/21 Added a call to PQsetNoticeProcessor() to catch pgsql "notices" and
2612 output them only if debugging. By default they are written stderr,
2613 apparently, which is not desirable.
2615 PH/22 Added Alain Williams' LDAP patch to support setting REFERRALS=off on
2618 JJ/01 exipick: added --reverse (and -R synonym), --random, --size, --sort and
2621 JJ/02 exipick: rewrote --help documentation to hopefully make more clear.
2623 PH/23 Made -oMaa and -oMt work with -bh and -bs to pretend the connection is
2624 authenticated or an ident call has been made. Suppress the default
2625 values for $authenticated_id and $authenticated_sender (but permit -oMai
2626 and -oMas) when testing with -bh.
2628 PH/24 Re-jigged the order of the tests in the default configuration so that the
2629 tests for valid domains and recipients precede the DNS black list and CSA
2630 tests, on the grounds that those ones are more expensive.
2632 PH/25 Exim was not testing for a space following SMTP commands such as EHLO
2633 that require one. Thus, EHLORHUBARB was interpreted as a valid command.
2634 This bug exists in every version of Exim that I still have, right back to
2637 PH/26 (n)wildlsearch lookups are documented as being done case-insensitively.
2638 However, an attempt to turn on case-sensitivity in a regex key by
2639 including (?-i) didn't work because the subject string was already
2640 lowercased, and the effects were non-intuitive. It turns out that a
2641 one-line patch can be used to allow (?-i) to work as expected.
2647 TF/01 Fix the add_header change below (4.61 PH/55) which had a bug that (amongst
2648 other effects) broke the use of negated acl sub-conditions.
2650 PH/01 ${readsocket now supports Internet domain sockets (modified John Jetmore
2653 PH/02 When tcp-wrappers is called from Exim, it returns only "deny" or "allow".
2654 "Deny" causes Exim to reject the incoming connection with a 554 error.
2655 Unfortunately, if there is a major crisis, such as a disk failure,
2656 tcp-wrappers gives "deny", whereas what one would like would be some
2657 kind of temporary error. A kludge has been added to help with this.
2658 Before calling hosts_ctl(), errno is set zero. If the result is "deny", a
2659 554 error is used if errno is still zero or contains ENOENT (which occurs
2660 if either of the /etc/hosts.{allow,deny} files is missing). Otherwise, a
2663 PH/03 Add -lutil to the default FreeBSD LIBS setting.
2665 PH/04 Change PH/19 for 4.61 was too wide. It should not be applied to host
2666 errors. Otherwise a message that provokes a temporary error (when other
2667 messages do not) can cause a whole host to time out.
2669 PH/05 Batch deliveries by appendfile and pipe transports did not work when the
2670 addresses were routed directly to files or pipes from a redirect router.
2671 File deliveries just didn't batch; pipe deliveries might have suffered
2674 PH/06 A failure to get a lock for a hints database would erroneously always say
2675 "Failed to get write lock", even when it was really a read lock.
2677 PH/07 The appendfile transport was creating MBX lock files with a fixed mode
2678 of 0600. This has been changed to use the value of the lockfile_mode
2679 option (which defaults to 0600).
2681 PH/08 Applied small patch from the Sieve maintainer.
2683 PH/09 If maildir_quota_directory_regex was set to exclude (say) the .Trash
2684 folder from quota calculations, a direct delivery into this folder messed
2685 up the contents of the maildirsize file. This was because the regex was
2686 used only to exclude .Trash (or whatever) when the size of the mailbox
2687 was calculated. There was no check that a delivery was happening into an
2688 excluded directory. This bug has been fixed by ignoring all quota
2689 processing for deliveries into excluded directories.
2691 PH/10 Added the maildirfolder_create_regex option to appendfile.
2697 PH/01 The code for finding all the local interface addresses on a FreeBSD
2698 system running IPv6 was broken. This may well have applied to all BSD
2699 systems, as well as to others that have similar system calls. The broken
2700 code found IPv4 interfaces correctly, but gave incorrect values for the
2701 IPv6 interfaces. In particular, ::1 was not found. The effect in Exim was
2702 that it would not match correctly against @[] and not recognize the IPv6
2705 PH/02 The ipliteral router was not recognizing addresses of the form user@
2706 [ipv6:....] because it didn't know about the "ipv6:" prefix.
2708 PH/03 Added disable_ipv6.
2710 PH/04 Changed $reply_address to use the raw form of the headers instead of the
2711 decoded form, because it is most often used to construct To: headers
2712 lines in autoreplies, and the decoded form may well be syntactically
2713 invalid. However, $reply_address has leading white space removed, and all
2714 newlines turned into spaces so that the autoreply transport does not
2717 PH/05 If group was specified without a user on a router, and no group or user
2718 was specified on a transport, the group from the router was ignored.
2720 PH/06 Increased the number of ACL variables to 20 of each type, and arranged
2721 for visible compile-time settings that can be used to change these
2722 numbers, for those that want even more. Backwards compatibility with old
2723 spool files has been maintained. However, going back to a previous Exim
2724 release will lost any variables that are in spool files.
2726 PH/07 Two small changes when running in the test harness: increase delay when
2727 passing a TCP/IP connection to a new process, in case the original
2728 process has to generate a bounce, and remove special handling of
2729 127.0.0.2 (sic), which is no longer necessary.
2731 PH/08 Changed debug output of dbfn_open() flags from numbers to names, so as to
2732 be the same on different OS.
2734 PH/09 Moved a debug statement in filter processing to avoid a race problem when
2737 JJ/01 exipick: fixed bug where -b (brief) output option showed "Vars:"
2738 whether --show-vars was specified or not
2740 JJ/02 exipick: Added support for new ACL variable spool format introduced
2743 PH/10 Fixed another bug related to PH/04 above: if an incoming message had a
2744 syntactically invalid From: or Reply-to: line, and a filter used this to
2745 generate an autoreply, and therefore failed to obtain an address for the
2746 autoreply, Exim could try to deliver to a non-existent relative file
2747 name, causing unrelated and misleading errors. What now happens is that
2748 it logs this as a hard delivery error, but does not attempt to create a
2751 PH/11 The exinext utility has a -C option for testing purposes, but although
2752 the given file was scanned by exinext itself; it wasn't being passed on
2753 when Exim was called.
2755 PH/12 In the smtp transport, treat an explicit ECONNRESET error the same as
2756 an end-of-file indication when reading a command response.
2758 PH/13 Domain literals for IPv6 were not recognized unless IPv6 support was
2759 compiled. In many other places in Exim, IPv6 addresses are always
2760 recognized, so I have changed this. It also means that IPv4 domain
2761 literals of the form [IPV4:n.n.n.n] are now always recognized.
2763 PH/14 When a uid/gid is specified for the queryprogram router, it cannot be
2764 used if the router is not running as root, for example, when verifying at
2765 ACL time, or when using -bh. The debugging output from this situation was
2766 non-existent - all you got was a failure to exec. I have made two
2769 (a) Failures to set uid/gid, the current directory, or a process leader
2770 in a subprocess such as that created by queryprogram now generate
2771 suitable debugging ouput when -d is set.
2773 (b) The queryprogram router detects when it is not running as root,
2774 outputs suitable debugging information if -d is set, and then runs
2775 the subprocess without attempting to change uid/gid.
2777 PH/15 Minor change to Makefile for building test_host (undocumented testing
2780 PH/16 As discussed on the list in Nov/Dec: Exim no longer looks at the
2781 additional section of a DNS packet that returns MX or SRV records.
2782 Instead, it always explicitly searches for A/AAAA records. This avoids
2783 major problems that occur when a DNS server includes only records of one
2784 type (A or AAAA) in an MX/SRV packet. A byproduct of this change has
2785 fixed another bug: if SRV records were looked up and the corresponding
2786 address records were *not* found in the additional section, the port
2787 values from the SRV records were lost.
2789 PH/17 If a delivery to a pipe, file, or autoreply was deferred, Exim was not
2790 using the correct key (the original address) when searching the retry
2791 rules in order to find which one to use for generating the retry hint.
2793 PH/18 If quota_warn_message contains a From: header, Exim now refrains from
2794 adding the default one. Similarly, if it contains a Reply-To: header, the
2795 errors_reply_to option, if set, is not used.
2797 PH/19 When calculating a retry time, Exim used to measure the "time since
2798 failure" by looking at the "first failed" field in the retry record. Now
2799 it does not use this if it is later than than the arrival time of the
2800 message. Instead it uses the arrival time. This makes for better
2801 behaviour in cases where some deliveries succeed, thus re-setting the
2802 "first failed" field. An example is a quota failure for a huge message
2803 when small messages continue to be delivered. Without this change, the
2804 "time since failure" will always be short, possible causing more frequent
2805 delivery attempts for the huge message than are intended.
2806 [Note: This change was subsequently modified - see PH/04 for 4.62.]
2808 PH/20 Added $auth1, $auth2, $auth3 to contain authentication data (as well as
2809 $1, $2, $3) because the numerical variables can be reset during some
2810 expansion items (e.g. "match"), thereby losing the authentication data.
2812 PH/21 Make -bV show the size of off_t variables so that the test suite can
2813 decide whether to run tests for quotas > 2G.
2815 PH/22 Test the values given for quota, quota_filecount, quota_warn_threshold,
2816 mailbox_size, and mailbox_filecount in the appendfile transport. If a
2817 filecount value is greater than 2G or if a quota value is greater than 2G
2818 on a system where the size of off_t is not greater than 4, a panic error
2821 PH/23 When a malformed item such as 1.2.3/24 appears in a host list, it can
2822 never match. The debug and -bh output now contains an explicit error
2823 message indicating a malformed IPv4 address or mask.
2825 PH/24 An host item such as 1.2.3.4/abc was being treated as the IP address
2826 1.2.3.4 without a mask. Now it is not recognized as an IP address, and
2827 PH/23 above applies.
2829 PH/25 Do not write to syslog when running in the test harness. The only
2830 occasion when this arises is a failure to open the main or panic logs
2831 (for which there is an explicit test).
2833 PH/26 Added the /no_tell option to "control=freeze".
2835 PH/27 If a host name lookup failed very early in a connection, for example, if
2836 the IP address matched host_lookup and the reverse lookup yielded a name
2837 that did not have a forward lookup, an error message of the form "no IP
2838 address found for host xxx.xxx.xxx (during SMTP connection from NULL)"
2839 could be logged. Now it outputs the IP address instead of "NULL".
2841 PH/28 An enabling patch from MH: add new function child_open_exim2() which
2842 allows the sender and the authenticated sender to be set when
2843 submitting a message from within Exim. Since child_open_exim() is
2844 documented for local_scan(), the new function should be too.
2846 PH/29 In GnuTLS, a forced expansion failure for tls_privatekey was not being
2847 ignored. In both GnuTLS and OpenSSL, an expansion of tls_privatekey that
2848 results in an empty string is now treated as unset.
2850 PH/30 Fix eximon buffer overflow bug (Bugzilla #73).
2852 PH/31 Added sender_verify_fail logging option.
2854 PH/32 In November 2003, the code in Exim that added an empty Bcc: header when
2855 needed by RFC 822 but not by RFC 2822 was commented out. I have now
2856 tidied the source and removed it altogether.
2858 PH/33 When a queue run was abandoned because the load average was too high, a
2859 log line was always written; now it is written only if the queue_run log
2860 selector is set. In addition, the log line for abandonment now contains
2861 information about the queue run such as the pid. This is always present
2862 in "start" and "stop" lines but was omitted from the "abandon" line.
2864 PH/34 Omit spaces between a header name and the colon in the error message that
2865 is given when verify = headers_syntax fails (if there are lots of them,
2866 the message gets confusing).
2868 PH/35 Change the default for dns_check_names_pattern to allow slashes within
2869 names, as there are now some PTR records that contain slashes. This check
2870 is only to protect against broken name servers that fall over on strange
2871 characters, so the fact that it applies to all lookups doesn't matter.
2873 PH/36 Now that the new test suite is complete, we can remove some of the
2874 special code in Exim that was needed for the old test suite. For example,
2875 sorting DNS records because real resolvers return them in an arbitrary
2876 order. The new test suite's fake resolver always returns records in the
2879 PH/37 When running in the test harness, use -odi for submitted messages (e.g.
2880 bounces) except when queue_only is set, to avoid logging races between
2881 the different processes.
2883 PH/38 Panic-die if .include specifies a non-absolute path.
2885 PH/39 A tweak to the "H" retry rule from its user.
2887 JJ/03 exipick: Removed parentheses from 'next' and 'last' calls that specified
2888 a label. They prevented compilation on older perls.
2890 JJ/04 exipick: Refactored code to prevent implicit split to @_ which caused
2891 a warning to be raised on newish perls.
2893 JJ/05 exipick: Fixed bug where -bpc always showed a count of all messages
2894 on queue. Changes to match documented behaviour of showing count of
2895 messages matching specified criteria.
2897 PH/40 Changed the default ident timeout from 30s to 5s.
2899 PH/41 Added support for the use of login_cap features, on those BSD systems
2900 that have them, for controlling the resources used by pipe deliveries.
2902 PH/42 The content-scanning code uses fopen() to create files in which to put
2903 message data. Previously it was not paying any attention to the mode of
2904 the files. Exim runs with umask(0) because the rest of the code creates
2905 files with open(), and sets the required mode explicitly. Thus, these
2906 files were ending up world-writeable. This was not a big issue, because,
2907 being within the spool directory, they were not world-accessible. I have
2908 created a function called modefopen, which takes an additional mode
2909 argument. It sets umask(777), creates the file, chmods it to the required
2910 mode, then resets the umask. All the relevant calls to fopen() in the
2911 content scanning code have been changed to use this function.
2913 PH/43 If retry_interval_max is set greater than 24 hours, it is quietly reset
2914 to 24 hours. This avoids potential overflow problems when processing G
2915 and H retry rules. I suspect nobody ever tinkers with this value.
2917 PH/44 Added STRIP_COMMAND=/usr/bin/strip to the FreeBSD Makefile.
2919 PH/45 When the plaintext authenticator is running as a client, the server's
2920 challenges are checked to ensure they are valid base64 strings. By
2921 default, the authentication attempt is cancelled if an invalid string is
2922 received. Setting client_ignore_invalid_base64 true ignores these errors.
2923 The decoded challenge strings are now placed in $auth1, $auth2, etc. as
2924 they are received. Thus, the responses can be made to depend on the
2925 challenges. If an invalid string is ignored, an empty string is placed in
2928 PH/46 Messages that are created by the autoreply transport now contains a
2929 References: header, in accordance with RFCs 2822 and 3834.
2931 PH/47 Added authenticated_sender_force to the smtp transport.
2933 PH/48 The ${prvs expansion was broken on systems where time_t was long long.
2935 PH/49 Installed latest patch from the Sieve maintainer.
2937 PH/50 When an Exim quota was set without a file count quota, and mailbox_size
2938 was also set, the appendfile transport was unnecessarily scanning a
2939 directory of message files (e.g. for maildir delivery) to find the count
2940 of files (along with the size), even though it did not need this
2941 information. It now does the scan only if it needs to find either the
2942 size of the count of files.
2944 PH/51 Added ${time_eval: to convert Exim time strings into seconds.
2946 PH/52 Two bugs concerned with error handling when the smtp transport is
2949 (i) Exim was not creating retry information for temporary errors given
2950 for individual recipients after the DATA command when the smtp transport
2951 was used in LMTP mode. This meant that they could be retried too
2952 frequently, and not timed out correctly.
2954 (ii) Exim was setting the flag that allows error details to be returned
2955 for LMTP errors on RCPT commands, but not for LMTP errors for individual
2956 recipients that were returned after the DATA command.
2958 PH/53 This is related to PH/52, but is more general: for any failing address,
2959 when detailed error information was permitted to be returned to the
2960 sender, but the error was temporary, then after the final timeout, only
2961 "retry timeout exceeded" was returned. Now it returns the full error as
2962 well as "retry timeout exceeded".
2964 PH/54 Added control=allow_auth_unadvertised, as it seems there are clients that
2965 do this, and (what is worse) MTAs that accept it.
2967 PH/55 Added the add_header modified to ACLs. The use of "message" with "warn"
2968 will now be deprecated.
2970 PH/56 New os.c-cygwin from the Cygwin maintainer.
2972 JJ/06 exipick: added --unsorted option to allow unsorted output in all output
2973 formats (previously only available in exim formats via -bpr, -bpru,
2974 and -bpra. Now also available in native and exiqgrep formats)
2976 JJ/07 exipick: added --freeze and --thaw options to allow faster interaction
2977 with very large, slow to parse queues
2979 JJ/08 exipick: added ! as generic prefix to negate any criteria format
2981 JJ/09 exipick: miscellaneous performance enhancements (~24% improvements)
2983 PH/57 Tidies in SMTP dialogue display in debug output: (i) It was not showing
2984 responses to authentication challenges, though it was showing the
2985 challenges; (ii) I've removed the CR characters from the debug output for
2988 PH/58 Allow for the insertion of a newline as well as a space when a string
2989 is turned into more than one encoded-word during RFC 2047 encoding. The
2990 Sieve code now uses this.
2992 PH/59 Added the following errors that can be detected in retry rules: mail_4xx,
2993 data_4xx, lost_connection, tls_required.
2995 PH/60 When a VRFY deferred or FAILED, the log message rather than the user
2996 message was being sent as an SMTP response.
2998 PH/61 Add -l and -k options to exicyclog.
3000 PH/62 When verifying, if an address was redirected to one new address, so that
3001 verification continued, and the new address failed or deferred after
3002 having set something in $address_data, the value of $address_data was not
3003 passed back to the ACL. This was different to the case when no
3004 redirection occurred. The value is now passed back in both cases.
3006 PH/63 Changed the macro HAVE_LOGIN_CAP (see PH/41 for this release above) to
3007 HAVE_SETCLASSRESOURCES because there are different APIs in use that all
3008 use login_cap.h, so on its own it isn't the distinguishing feature. The
3009 new name refers directly to the setclassresources() function.
3011 PH/65 Added configuration files for NetBSD3.
3013 PH/66 Updated OS/Makefile-HP-UX for gcc 4.1.0 with HP-UX 11.
3015 PH/67 Fixed minor infelicity in the sorting of addresses to ensure that IPv6
3016 is preferred over IPv4.
3018 PH/68 The bounce_return_message and bounce_return_body options were not being
3019 honoured for bounces generated during the reception of non-SMTP messages.
3020 In particular, this applied to messages rejected by the ACL. This bug has
3021 been fixed. However, if bounce_return_message is true and bounce_return_
3022 body is false, the headers that are returned for a non-SMTP message
3023 include only those that have been read before the error was detected.
3024 (In the case of an ACL rejection, they have all been read.)
3026 PH/69 The HTML version of the specification is now built in a directory called
3027 spec_html instead of spec.html, because the latter looks like a path with
3028 a MIME-type, and this confuses some software.
3030 PH/70 Catch two compiler warnings in sieve.c.
3032 PH/71 Fixed an obscure and subtle bug (thanks Alexander & Matthias). The
3033 function verify_get_ident() calls ip_connect() to connect a socket, but
3034 if the "connect()" function timed out, ip_connect() used to close the
3035 socket. However, verify_get_ident() also closes the socket later, and in
3036 between Exim writes to the log, which may get opened at this point. When
3037 the socket was closed in ip_connect(), the log could get the same file
3038 descriptor number as the socket. This naturally causes chaos. The fix is
3039 not to close the socket in ip_connect(); the socket should be closed by
3040 the function that creates it. There was only one place in the code where
3041 this was missing, in the iplookup router, which I don't think anybody now
3042 uses, but I've fixed it anyway.
3044 PH/72 Make dns_again_means_nonexist apply to lookups using gethostbyname() as
3045 well as to direct DNS lookups. Otherwise the handling of names in host
3046 lists is inconsistent and therefore confusing.
3052 PH/01 Two changes to the default runtime configuration:
3054 (1) Move the checks for relay_from_hosts and authenticated clients from
3055 after to before the (commented out) DNS black list checks.
3057 (2) Add control=submission to the relay_from_hosts and authenticated
3058 clients checks, on the grounds that messages accepted by these
3059 statements are most likely to be submissions.
3061 PH/02 Several tidies to the handling of ${prvs and ${prvscheck:
3063 (1) Generate an error if the third argument for the ${prvs expansion is
3066 (2) Treat a missing third argument of ${prvscheck as if it were an empty
3069 (3) Reset the variables that are obtained from the first argument of
3070 ${prvscheck and used in the second argument before leaving the code,
3071 because their memory is reclaimed, so using them afterwards may do
3074 (4) Tidy up the code for expanding the arguments of ${prvscheck one by
3075 one (it's much easier than Tom thought :-).
3077 (5) Because of (4), we can now allow for the use of $prvscheck_result
3078 inside the third argument.
3080 PH/03 For some reason, the default setting of PATH when running a command from
3081 a pipe transport was just "/usr/bin". I have changed it to
3084 PH/04 SUPPORT_TRANSLATE_IP_ADDRESS and MOVE_FROZEN_MESSAGES did not cause
3085 anything to be listed in the output from -bV.
3087 PH/05 When a filter generated an autoreply, the entire To: header line was
3088 quoted in the delivery log line, like this:
3090 => >A.N.Other <ano@some.domain> <original@ddress> ...
3092 This has been changed so that it extracts the operative address. There
3093 may be more than one such address. If so, they are comma-separated, like
3096 => >ano@some.domain,ona@other.domain <original@ddress> ...
3098 PH/06 When a client host used a correct literal IP address in a HELO or EHLO
3099 command, (for example, EHLO [1.2.3.4]) and the client's IP address was
3100 not being looked up in the rDNS to get a host name, Exim was showing the
3101 IP address twice in Received: lines, even though the IP addresses were
3102 identical. For example:
3104 Received: from [1.2.3.4] (helo=[1.2.3.4])
3106 However, if the real host name was known, it was omitting the HELO data
3107 if it matched the actual IP address. This has been tidied up so that it
3108 doesn't show the same IP address twice.
3110 PH/07 When both +timestamp and +memory debugging was on, the value given by
3111 $tod_xxx expansions could be wrong, because the tod_stamp() function was
3112 called by the debug printing, thereby overwriting the timestamp buffer.
3113 Debugging no longer uses the tod_stamp() function when +timestamp is set.
3115 PH/08 When the original message was included in an autoreply transport, it
3116 always said "this is a copy of the message, including all the headers",
3117 even if body_only or headers_only was set. It now gives an appropriate
3120 PH/09 Applied a patch from the Sieve maintainer which:
3122 o fixes some comments
3123 o adds the (disabled) notify extension core
3124 o adds some debug output for the result of if/elsif tests
3125 o points to the current vacation draft in the documentation
3126 and documents the missing references header update
3130 o fixes a bug in processing the envelope test (when testing
3131 multiple envelope elements, the last element determinted the
3134 PH/10 Exim was violating RFC 3834 ("Recommendations for Automatic Responses to
3135 Electronic Mail") by including:
3137 Auto-submitted: auto-generated
3139 in the messages that it generates (bounce messages and others, such as
3140 warnings). In the case of bounce messages for non-SMTP mesages, there was
3141 also a typo: it was using "Auto_submitted" (underscore instead of
3142 hyphen). Since every message generated by Exim is necessarily in response
3143 to another message, thes have all been changed to:
3145 Auto-Submitted: auto-replied
3147 in accordance with these statements in the RFC:
3149 The auto-replied keyword:
3151 - SHOULD be used on messages sent in direct response to another
3152 message by an automatic process,
3154 - MUST NOT be used on manually-generated messages,
3156 - MAY be used on Delivery Status Notifications (DSNs) and Message
3157 Disposition Notifications (MDNs),
3159 - MUST NOT be used on messages generated by automatic or periodic
3160 processes, except for messages which are automatic responses to
3163 PH/11 Added "${if def:sender_address {(envelope-from <$sender_address>)\n\t}}"
3164 to the default Received: header definition.
3166 PH/12 Added log selector acl_warn_skipped (default on).
3168 PH/13 After a successful wildlsearch lookup, discard the values of numeric
3169 variables because (a) they are in the wrong storage pool and (b) even if
3170 they were copied, it wouldn't work properly because of the caching.
3172 PH/14 Add check_rfc2047_length to disable enforcement of RFC 2047 length
3173 checking when decoding. Apparently there are clients that generate
3174 overlong encoded strings. Why am I not surprised?
3176 PH/15 If the first argument of "${if match_address" was not empty, but did not
3177 contain an "@" character, Exim crashed. Now it writes a panic log message
3178 and treats the condition as false.
3180 PH/16 In autoreply, treat an empty string for "once" the same as unset.
3182 PH/17 A further patch from the Sieve maintainer: "Introduce the new Sieve
3183 extension "envelope-auth". The code is finished and in agreement with
3184 other implementations, but there is no documentation so far and in fact,
3185 nobody wrote the draft yet. This extension is currently #undef'ed, thus
3186 not changing the active code.
3188 Print executed "if" and "elsif" statements when debugging is used. This
3189 helps a great deal to understand what a filter does.
3191 Document more things not specified clearly in RFC3028. I had all this
3192 sorted out, when out of a sudden new issues came to my mind. Oops."
3194 PH/18 Exim was not recognizing the "net-" search type prefix in match_ip lists
3197 PH/19 Exim expands the IPv6 address given to -bh to its full non-abbreviated
3198 canonical form (as documented). However, after a host name lookup from
3199 the IP address, check_host() was doing a simple string comparison with
3200 addresses acquired from the DNS when checking that the found name did
3201 have the original IP as one of its addresses. Since any found IPv6
3202 addresses are likely to be in abbreviated form, the comparison could
3203 fail. Luckily, there already exists a function for doing the comparison
3204 by converting both addresses to binary, so now that is used instead of
3205 the text comparison.
3207 PH/20 There was another similar case to PH/19, when a complete host name was
3208 given in a host list; looking up its IP address could give an abbreviated
3209 form, whereas the current host's name might or might not be abbreviated.
3210 The same fix has been applied.
3216 PH/01 The ${base62: operator adjusted itself to base 36 when BASE_62 was
3217 set to 36 (for Darwin and Cygwin), but the ${base62d: operator did not.
3220 PH/02 Two minor problems detected in Cygwin: the os.{c,h} files had lost */ on
3221 the CVS lines, and there was a missing #if HAVE_IPV6 in host.c.
3223 PH/03 Typo: missing ".o" in src/pcre/Makefile.
3225 PH/04 Tighten up "personal" tests: Instead of testing for any "List-"
3226 header line, restrict the check to what is listed in RFCs 2369 and 2929.
3227 Also, for "Auto-Submitted", treat anything other than "no" as
3228 non-personal, in accordance with RFC 3834. (Previously it treated
3229 anything starting "auto-" as non-personal.)
3231 TF/01 The control=submission/name=... option had a problem with syntax
3232 errors if the name included a slash character. The /name= option
3233 now slurps the rest of the string, so it can include any characters
3234 but it must come last in the list of options (after /sender_retain
3237 PH/05 Some modifications to the interface to the fake nameserver for the new
3245 TK/01 Added the "success_on_redirect" address verification option. See
3246 NewStuff for rationale and an example.
3248 PH/01 Added support for SQLite, basic code supplied by David Woodhouse.
3250 PH/02 Patch to exigrep to allow it to work on syslog lines.
3252 PH/03 When creating an mbox file for a virus/spam scan, use fseek() instead of
3253 fread() to skip over the body file's header line, because in Cygwin the
3254 header line is locked and is inaccessible.
3256 PH/04 Added $message_exim_id, ultimately to replace $message_id (they will both
3257 co-exist for some time) to make it clear that it is the Exim ID that is
3258 referenced, not the Message-ID: header line.
3260 PH/05 Replaced all Tom's calls to snprintf() with calls to the internal
3261 string_format() function, because snprintf() does not exist on all
3264 PH/06 The use of forbid_filter_existstest now also locks out the use of the
3265 ${stat: expansion item.
3267 PH/07 Changed "SMTP protocol violation: synchronization error" into "SMTP
3268 protocol synchronization error", to keep the pedants happy.
3270 PH/08 Arrange for USE_INET_NTOA_FIX to be set in config.h for AIX systems as
3271 well as for IRIX systems, when gcc is being used. See the host.c source
3274 PH/09 Installed latest Cygwin configuration files from the Cygwin maintainer.
3276 PH/10 Named domain lists were not working if used in a queue_smtp_domains
3279 PH/11 Added support for the IGNOREQUOTA extension to LMTP, both to the lmtp
3280 transport and to the smtp transport in LMTP mode.
3282 TK/02 Remove one case of BASE64 error detection FTTB (undocumented anyway).
3284 PH/12 There was a missing call to search_tidyup() before the fork() in rda.c to
3285 run a filter in a subprocess. This could lead to confusion in subsequent
3286 lookups in the parent process. There should also be a search_tidyup() at
3287 the end of the subprocess.
3289 PH/13 Previously, if "verify = helo" was set in an ACL, the condition was true
3290 only if the host matched helo_try_verify_hosts, which caused the
3291 verification to occur when the EHLO/HELO command was issued. The ACL just
3292 tested the remembered result. Now, if a previous verification attempt has
3293 not happened, "verify = helo" does it there and then.
3295 JJ/01 exipick: added $message_exim_id variable (see 4.53-PH/04)
3297 TK/03 Fix log output including CR from clamd.
3299 PH/14 A reference to $reply_address when Reply-to: was empty and From: did not
3300 exist provoked a memory error which could cause a segfault.
3302 PH/15 Installed PCRE 6.2
3304 PH/17 Defined BIND_8_COMPAT in the Darwin os.h file.
3306 PH/18 Reversed 4.52/PH/17 because the HP-UX user found it wasn't the cause
3307 of the problem. Specifically, suggested +O2 rather than +O1 for the
3310 PH/19 Added sqlite_lock_timeout option (David Woodhouse's patch).
3312 PH/20 If a delivery was routed to a non-standard port by means of an SRV
3313 record, the port was not correctly logged when the outgoing_port log
3314 selector was set (it logged the transort's default port).
3316 PH/21 Added support for host-specific ports to manualroute, queryprogram,
3317 fallback_hosts, and "hosts" in the smtp transport.
3319 PH/22 If the log selector "outgoing_port" is set, the port is now also given on
3320 host errors such as "Connection refused".
3322 PH/23 Applied a patch to fix problems with exim-4.52 while doing radius
3323 authentication with radiusclient 0.4.9:
3325 - Error returned from rc_read_config was caught wrongly
3326 - Username/password not passed on to radius server due to wrong length.
3328 The presumption is that some radiusclient API changes for 4.51/PH/17
3329 were not taken care of correctly. The code is still untested by me (my
3330 Linux distribution still has 0.3.2 of radiusclient), but it was
3331 contributed by a Radius user.
3333 PH/24 When doing a callout, the value of $domain wasn't set correctly when
3334 expanding the "port" option of the smtp transport.
3336 TK/04 MIME ACL: Fix buffer underrun that occurs when EOF condition is met
3337 while reading a MIME header. Thanks to Tom Hughes for a patch.
3339 PH/24 Include config.h inside local_scan.h so that configuration settings are
3342 PH/25 Make $smtp_command_argument available after all SMTP commands. This means
3343 that in an ACL for RCPT (for example), you can examine exactly what was
3346 PH/26 Exim was recognizing IPv6 addresses of the form [IPv6:....] in EHLO
3347 commands, but it was not correctly comparing the address with the actual
3348 client host address. Thus, it would show the EHLO address in Received:
3349 header lines when this was not necessary.
3351 PH/27 Added the % operator to ${eval:}.
3353 PH/28 Exim tries to create and chdir to its spool directory when it starts;
3354 it should be ignoring failures (because with -C, for example, it has lost
3355 privilege). It wasn't ignoring creation failures other than "already
3358 PH/29 Added "crypteq" to the list of supported features that Exim outputs when
3361 PH/30 Fixed (presumably very longstanding) bug in exim_dbmbuild: if it failed
3362 because an input line was too long, either on its own, or by virtue of
3363 too many continuations, the temporary file was not being removed, and the
3364 return code was incorrect.
3366 PH/31 Missing "BOOL" in function definition in filtertest.c.
3368 PH/32 Applied Sieve patches from the maintainer.
3370 TK/05 Domainkeys: Accomodate for a minor API change in libdomainkeys 0.67.
3372 PH/33 Added "verify = not_blind".
3374 PH/34 There are settings for CHOWN_COMMAND and MV_COMMAND that can be used in
3375 Local/Makefile (with some defaults set). These are used in built scripts
3376 such as exicyclog, but they have never been used in the exim_install
3377 script (though there are many overriding facilities there). I have
3378 arranged that the exim_install script now takes note of these two
3381 PH/35 Installed configuration files for Dragonfly.
3383 PH/36 When a locally submitted message by a trusted user did not contain a
3384 From: header, and the sender address was obtained from -f or from an SMTP
3385 MAIL command, and the trusted user did not use -F to supply a sender
3386 name, $originator_name was incorrectly used when constructing a From:
3387 header. Furthermore, $originator_name was used for submission mode
3388 messages from external hosts without From: headers in a similar way,
3389 which is clearly wrong.
3391 PH/37 Added control=suppress_local_fixups.
3393 PH/38 When log_selector = +received_sender was set, and the addition of the
3394 sender made the log line's construction buffer exactly full, or one byte
3395 less than full, an overflow happened when the terminating "\n" was
3398 PH/39 Added a new log selector, "unknown_in_list", which provokes a log entry
3399 when the result of a list match is failure because a DNS lookup failed.
3401 PH/40 RM_COMMAND is now used in the building process.
3403 PH/41 Added a "distclean" target to the top-level Makefile; it deletes all
3404 the "build-* directories that it finds.
3406 PH/42 (But a TF fix): In a domain list, Exim incorrectly matched @[] if the IP
3407 address in a domain literal was a prefix of an interface address.
3409 PH/43 (Again a TF fix): In the dnslookup router, do not apply widen_domains
3410 when verifying a sender address, unless rewrite_headers is false.
3412 PH/44 Wrote a long comment about why errors_to addresses are verified as
3413 recipients, not senders.
3415 TF/01 Add missing LIBS=-lm to OS/Makefile-OpenBSD which was overlooked when
3416 the ratelimit ACL was added.
3418 PH/45 Added $smtp_command for the full command (cf $smtp_command_argument).
3420 PH/46 Added extra information about PostgreSQL errors to the error string.
3422 PH/47 Added an interface to a fake DNS resolver for use by the new test suite,
3423 avoiding the need to install special zones in a real server. This is
3424 backwards compatible; if it can't find the fake resolver, it drops back.
3425 Thus, both old and new test suites can be run.
3427 TF/02 Added util/ratelimit.pl
3429 TF/03 Minor fix to the ratelimit code to improve its behaviour in case the
3430 clock is set back in time.
3432 TF/04 Fix the ratelimit support in exim_fixdb. Patch provided by Brian
3433 Candler <B.Candler@pobox.com>.
3435 TF/05 The fix for PH/43 was not completely correct; widen_domains is always
3436 OK for addresses that are the result of redirections.
3438 PH/48 A number of further additions for the benefit of the new test suite,
3439 including a fake gethostbyname() that interfaces to the fake DNS resolver
3442 TF/06 The fix for widen_domains has also been applied to qualify_single and
3443 search_parents which are the other dnslookup options that can cause
3446 PH/49 Michael Haardt's randomized retrying, but as a separate retry parameter
3449 PH/50 Make never_users, trusted_users, admin_groups, trusted_groups expandable.
3451 TF/07 Exim produced the error message "an SRV record indicated no SMTP
3452 service" if it encountered an MX record with an empty target hostname.
3453 The message is now "an MX or SRV record indicated no SMTP service".
3455 TF/08 Change PH/13 introduced the possibility that verify=helo may defer,
3456 if the DNS of the sending site is misconfigured. This is quite a
3457 common situation. This change restores the behaviour of treating a
3458 helo verification defer as a failure.
3460 PH/51 If self=fail was set on a router, the bounce message did not include the
3461 actual error message.
3467 TF/01 Added support for Client SMTP Authorization. See NewStuff for details.
3469 PH/01 When a transport filter timed out in a pipe delivery, and the pipe
3470 command itself ended in error, the underlying message about the transport
3471 filter timeout was being overwritten with the pipe command error. Now the
3472 underlying error message should be appended to the second error message.
3474 TK/01 Fix poll() being unavailable on Mac OSX 10.2.
3476 PH/02 Reduce the amount of output that "make" produces by default. Full output
3477 can still be requested.
3479 PH/03 The warning log line about a condition test deferring for a "warn" verb
3480 was being output only once per connection, rather than after each
3481 occurrence (because it was using the same function as for successful
3482 "warn" verbs). This seems wrong, so I have changed it.
3484 TF/02 Two buglets in acl.c which caused Exim to read a few bytes of memory that
3485 it should not have, which might have caused a crash in the right
3486 circumstances, but probably never did.
3488 PH/04 Installed a modified version of Tony Finch's patch to make submission
3489 mode fix the return path as well as the Sender: header line, and to
3490 add a /name= option so that you can make the user's friendly name appear
3493 TF/03 Added the control = fakedefer ACL modifier.
3495 TF/04 Added the ratelimit ACL condition. See NewStuff for details. Thanks to
3496 Mark Lowes for thorough testing.
3498 TK/02 Rewrote SPF support to work with libspf2 versions >1.2.0.
3500 TK/03 Merged latest SRS patch from Miles Wilton.
3502 PH/05 There's a shambles in IRIX6 - it defines EX_OK in unistd.h which conflicts
3503 with the definition in sysexits.h (which is #included earlier).
3504 Fortunately, Exim does not actually use EX_OK. The code used to try to
3505 preserve the sysexits.h value, by assumimg that macro definitions were
3506 scanned for macro replacements. I have been disabused of this notion,
3507 so now the code just undefines EX_OK before #including unistd.h.
3509 PH/06 There is a timeout for writing blocks of data, set by, e.g. data_timeout
3510 in the smtp transport. When a block could not be written in a single
3511 write() function, the timeout was being re-applied to each part-write.
3512 This seems wrong - if the receiver was accepting one byte at a time it
3513 would take for ever. The timeout is now adjusted when this happens. It
3514 doesn't have to be particularly precise.
3516 TK/04 Added simple SPF lookup method in EXPERIMENTAL_SPF. See NewStuff for
3517 details. Thanks to Chris Webb <chris@arachsys.com> for the patch!
3519 PH/07 Added "fullpostmaster" verify option, which does a check to <postmaster>
3520 without a domain if the check to <postmaster@domain> fails.
3522 SC/01 Eximstats: added -xls and the ability to specify output files
3523 (patch written by Frank Heydlauf).
3525 SC/02 Eximstats: use FileHandles for outputing results.
3527 SC/03 Eximstats: allow any combination of xls, txt, and html output.
3529 SC/04 Eximstats: fixed display of large numbers with -nvr option
3531 SC/05 Eximstats: fixed merging of reports with empty tables.
3533 SC/06 Eximstats: added the -include_original_destination flag
3535 SC/07 Eximstats: removed tabs and trailing whitespace.
3537 TK/05 Malware: Improve on aveserver error handling. Patch from Alex Miller.
3539 TK/06 MBOX spool code: Add real "From " MBOX separator line
3540 so the .eml file is really in mbox format (even though
3541 most programs do not really care). Patch from Alex Miller.
3543 TK/07 MBOX spool code: Add X-Envelope-From: and X-Envelope-To: headers.
3544 The latter is generated from $received_to and is only set if the
3545 message has one envelope recipient. SA can use these headers,
3546 obviously out-of-the-box. Patch from Alex Miller.
3548 PH/08 The ${def test on a variable was returning false if the variable's
3549 value was "0", contrary to what the specification has always said!
3550 The result should be true unless the variable is empty.
3552 PH/09 The syntax error of a character other than { following "${if
3553 def:variable_name" (after optional whitespace) was not being diagnosed.
3554 An expansion such as ${if def:sender_ident:{xxx}{yyy}} in which an
3555 accidental colon was present, for example, could give incorrect results.
3557 PH/10 Tidied the code in a number of places where the st_size field of a stat()
3558 result is used (not including appendfile, where other changes are about
3561 PH/11 Upgraded appendfile so that quotas larger than 2G are now supported.
3562 This involved changing a lot of size variables from int to off_t. It
3563 should work with maildirs and everything.
3565 TK/08 Apply fix provided by Michael Haardt to prevent deadlock in case of
3566 spamd dying while we are connected to it.
3568 TF/05 Fixed a ${extract error message typo reported by Jeremy Harris
3571 PH/12 Applied Alex Kiernan's patch for the API change for the error callback
3572 function for BDB 4.3.
3574 PH/13 Changed auto_thaw such that it does not apply to bounce messages.
3576 PH/14 Imported PCRE 6.0; this was more than just a trivial operation because
3577 the sources for PCRE have been re-arranged and more files are now
3580 PH/15 The code I had for printing potentially long long variables in PH/11
3581 above was not the best (it lost precision). The length of off_t variables
3582 is now inspected at build time, and an appropriate printing format (%ld
3583 or %lld) is chosen and #defined by OFF_T_FMT. We also define LONGLONG_T
3584 to be "long long int" or "long int". This is needed for the internal
3585 formatting function string_vformat().
3587 PH/16 Applied Matthew Newton's patch to exicyclog: "If log_file_path is set in
3588 the configuration file to be ":syslog", then the script "guesses" where
3589 the logs files are, rather than using the compiled in default. In our
3590 case the guess is not the same as the compiled default, so the script
3591 suddenly stopped working when I started to use syslog. The patch checks
3592 to see if log_file_path is "". If so, it attempts to read it from exim
3593 with no configuration file to get the compiled in version, before it
3594 falls back to the previous guessing code."
3596 TK/09 Added "prvs" and "prvscheck" expansion items. These help a lot with
3597 implementing BATV in an Exim configuration. See NewStuff for the gory
3600 PH/17 Applied Michael Haardt's patch for HP-UX, affecting only the os.h and
3601 Makefile that are specific to HP-UX.
3603 PH/18 If the "use_postmaster" option was set for a recipient callout together
3604 with the "random" option, the postmaster address was used as the MAIL
3605 FROM address for the random test, but not for the subsequent recipient
3606 test. It is now used for both.
3608 PH/19 Applied Michael Haardt's patch to update Sieve to RFC3028bis. "The
3609 patch removes a few documentation additions to RFC 3028, because the
3610 latest draft now contains them. It adds the new en;ascii-case comparator
3611 and a new error check for 8bit text in MIME parts. Comparator and
3612 require names are now matched exactly. I enabled the subaddress
3613 extension, but it is not well tested yet (read: it works for me)."
3615 PH/20 Added macros for time_t as for off_t (see PH/15 above) and used them to
3616 rework some of the code of TK/09 above to avoid the hardwired use of
3617 "%lld" and "long long". Replaced the call to snprintf() with a call to
3620 PH/21 Added some other messages to those in 4.51/PH/42, namely "All relevant MX
3621 records point to non-existent hosts", "retry timeout exceeded", and
3622 "retry time not reached for any host after a long failure period".
3624 PH/22 Fixed some oversights/typos causing bugs when Exim is compiled with
3625 experimental DomainKeys support:
3627 (1) The filter variables $n0-$n9 and $sn0-$sn9 were broken.
3628 (2) On an error such as an illegally used "control", the wrong name for
3629 the control was given.
3631 These problems did NOT occur unless DomainKeys support was compiled.
3633 PH/23 Added daemon_startup_retries and daemon_startup_sleep.
3635 PH/24 Added ${if match_ip condition.
3637 PH/25 Put debug statements on either side of calls to EXIM_DBOPEN() for hints
3638 databases so that it will be absolutely obvious if a crash occurs in the
3639 DB library. This is a regular occurrence (often caused by mis-matched
3642 PH/26 Insert a lot of missing (void) casts for functions such as chown(),
3643 chmod(), fcntl(), sscanf(), and other functions from stdio.h. These were
3644 picked up on a user's system that detects such things. There doesn't seem
3645 to be a gcc warning option for this - only an attribute that has to be
3646 put on the function's prototype. It seems that in Fedora Core 4 they have
3647 set this on a number of new functions. No doubt there will be more in due
3650 PH/27 If a dnslookup or manualroute router is set with verify=only, it need not
3651 specify a transport. However, if an address that was verified by such a
3652 router was the subject of a callout, Exim crashed because it tried to
3653 read the rcpt_include_affixes from the non-existent transport. Now it
3654 just assumes that the setting of that option is false. This bug was
3655 introduced by 4.51/PH/31.
3657 PH/28 Changed -d+all to exclude +memory, because that information is very
3658 rarely of interest, but it makes the output a lot bigger. People tend to
3659 do -d+all out of habit.
3661 PH/29 Removed support for the Linux-libc5 build, as it is obsolete and the
3662 code in os-type was giving problems when libc.so lives in lib64, like on
3665 PH/30 Exim's DNS code uses the original T_xxx names for DNS record times. These
3666 aren't the modern standard, and it seems that some systems' include files
3667 don't always have them. Exim was already checking for some of the newer
3668 ones like T_AAAA, and defining it itself. I've added checks for all the
3669 record types that Exim uses.
3671 PH/31 When using GnuTLS, if the parameters cache file did not exist, Exim was
3672 not automatically generating a new one, as it is supposed to. This
3673 prevented TLS from working. If the file did exist, but contained invalid
3674 data, a new version was generated, as expected. It was only the case of a
3675 non-existent file that was broken.
3677 TK/10 Domainkeys: Fix a bug in verification that caused a crash in conjunction
3678 with a change in libdomainkeys > 0.64.
3680 TK/11 Domainkeys: Change the logic how the "testing" policy flag is retrieved
3681 from DNS. If the selector record carries the flag, it now has
3682 precedence over the domain-wide flag.
3684 TK/12 Cleared some compiler warnings related to SPF, SRS and DK code.
3686 PH/32 In mua_wrapper mode, if an smtp transport configuration error (such as
3687 the use of a port name that isn't defined in /etc/services) occurred, the
3688 message was deferred as in a normal delivery, and thus remained on the
3689 spool, instead of being failed because of the mua_wrapper setting. This
3690 is now fixed, and I tidied up some of the mua_wrapper messages at the
3693 SC/08 Eximstats: whilst parsing the mainlog(s), store information about
3694 the messages in a hash of arrays rather than using individual hashes.
3695 This is a bit cleaner and results in dramatic memory savings, albeit
3696 at a slight CPU cost.
3698 SC/09 Eximstats: added the -show_rt<list> and the -show_dt<list> flags
3699 as requested by Marc Sherman.
3701 SC/10 Eximstats: added histograms for user specified patterns as requested
3704 SC/11 Eximstats: v1.43 - bugfix for pattern histograms with -h0 specified.
3706 PH/33 Patch from the Cygwin maintainer to add "b" to all occurences of
3707 fopen() in the content-scanning modules that did not already have it.
3713 TK/01 Added Yahoo DomainKeys support via libdomainkeys. See
3714 doc/experimental-spec.txt for details. (http://domainkeys.sf.net)
3716 TK/02 Fix ACL "control" statement not being available in MIME ACL.
3718 TK/03 Fix ACL "regex" condition not being available in MIME ACL.
3720 PH/01 Installed a patch from the Sieve maintainer that allows -bf to be used
3721 to test Sieve filters that use "vacation".
3723 PH/02 Installed a slightly modified version of Nikos Mavrogiannopoulos' patch
3724 that changes the way the GnuTLS parameters are stored in the cache file.
3725 The new format can be generated externally. For backward compatibility,
3726 if the data in the cache doesn't make sense, Exim assumes it has read an
3727 old-format file, and it generates new data and writes a new file. This
3728 means that you can't go back to an older release without removing the
3731 PH/03 A redirect router that has both "unseen" and "one_time" set does not
3732 work if there are any delivery delays because "one_time" forces the
3733 parent to be marked "delivered", so its unseen clone is never tried
3734 again. For this reason, Exim now forbids the simultaneous setting of
3737 PH/04 Change 4.11/85 fixed an obscure bug concerned with addresses that are
3738 redirected to themselves ("homonym" addresses). Read the long ChangeLog
3739 entry if you want to know the details. The fix, however, neglected to
3740 consider the case when local delivery batching is involved. The test for
3741 "previously delivered" was not happening when checking to see if an
3742 address could be batched with a previous (undelivered) one; under
3743 certain circumstances this could lead to multiple deliveries to the same
3746 PH/05 Renamed the macro SOCKLEN_T as EXIM_SOCKLEN_T because AIX uses SOCKLEN_T
3747 in its include files, and this causes problems building Exim.
3749 PH/06 A number of "verify =" ACL conditions have no options (e.g. verify =
3750 header_syntax) but Exim was just ignoring anything given after a slash.
3751 In particular, this caused confusion with an attempt to use "verify =
3752 reverse_host_lookup/defer_ok". An error is now given when options are
3753 supplied for verify items that do not have them. (Maybe reverse_host_
3754 lookup should have a defer_ok option, but that's a different point.)
3756 PH/07 Increase the size of the buffer for incoming SMTP commands from 512 (as
3757 defined by RFC 821) to 2048, because there were problems with some AUTH
3758 commands, and RFC 1869 says the size should be increased for extended
3759 SMTP commands that take arguments.
3761 PH/08 Added ${dlfunc dynamically loaded function for expansion (code from Tony
3764 PH/09 Previously, an attempt to use ${perl when it wasn't compiled gave an
3765 "unknown" error; now it says that the functionality isn't in the binary.
3767 PH/10 Added a nasty fudge to try to recognize and flatten LDAP passwords in
3768 an address' error message when a string expansion fails (syntax or
3769 whatever). Otherwise the password may appear in the log. Following change
3770 PH/42 below, there is no longer a chance of it appearing in a bounce
3773 PH/11 Installed exipick version 20050225.0 from John Jetmore.
3775 PH/12 If the last host in a fallback_hosts list was multihomed, only the first
3776 of its addresses was ever tried. (Bugzilla bug #2.)
3778 PH/13 If "headers_add" in a transport didn't end in a newline, Exim printed
3779 the result incorrectly in the debug output. (It correctly added a newline
3780 to what was transported.)
3782 TF/01 Added $received_time.
3784 PH/14 Modified the default configuration to add an acl_smtp_data ACL, with
3785 commented out examples of how to interface to a virus scanner and to
3786 SpamAssassin. Also added commented examples of av_scanner and
3787 spamd_address settings.
3789 PH/15 Further to TK/02 and TK/03 above, tidied up the tables of what conditions
3790 and controls are allowed in which ACLs. There were a couple of minor
3791 errors. Some of the entries in the conditions table (which is a table of
3792 where they are NOT allowed) were getting very unwieldy; rewrote them as a
3793 negation of where the condition IS allowed.
3795 PH/16 Installed updated OS/os.c-cygwin from the Cygwin maintainer.
3797 PH/17 The API for radiusclient changed at release 0.4.0. Unfortunately, the
3798 header file does not have a version number, so I've had to invent a new
3799 value for RADIUS_LIB_TYPE, namely "RADIUSCLIENTNEW" to request the new
3800 API. The code is untested by me (my Linux distribution still has 0.3.2 of
3801 radiusclient), but it was contributed by a Radius user.
3803 PH/18 Installed Lars Mainka's patch for the support of CRL collections in
3804 files or directories, for OpenSSL.
3806 PH/19 When an Exim process that is running as root has to create an Exim log
3807 file, it does so in a subprocess that runs as exim:exim so as to get the
3808 ownership right at creation (otherwise, other Exim processes might see
3809 the file with the wrong ownership). There was no test for failure of this
3810 fork() call, which would lead to the process getting stuck as it waited
3811 for a non-existent subprocess. Forks do occasionally fail when resources
3812 run out. I reviewed all the other calls to fork(); they all seem to check
3815 PH/20 When checking for unexpected SMTP input at connect time (before writing
3816 the banner), Exim was not dealing correctly with a non-positive return
3817 from the read() function. If the client had disconnected by this time,
3818 the result was a log entry for a synchronization error with an empty
3819 string after "input=" when read() returned zero. If read() returned -1
3820 (an event I could not check), uninitialized data bytes were printed.
3821 There were reports of junk text (parts of files, etc) appearing after
3824 PH/21 Added acl_not_smtp_mime to allow for MIME scanning for non-SMTP messages.
3826 PH/22 Added support for macro redefinition, and (re)definition in between
3827 driver and ACL definitions.
3829 PH/23 The cyrus_sasl authenticator was expanding server_hostname, but then
3830 forgetting to use the resulting value; it was using the unexpanded value.
3832 PH/24 The cyrus_sasl authenticator was advertising mechanisms for which it
3833 hadn't been configured. The fix is from Juergen Kreileder, who
3834 understands it better than I do:
3836 "Here's what I see happening with three configured cyrus_sasl
3837 authenticators configured (plain, login, cram-md5):
3839 On startup auth_cyrus_sasl_init() gets called for each of these.
3840 This means three calls to sasl_listmech() without a specified mech_list.
3841 => SASL tests which mechs of all available mechs actually work
3842 => three warnings about OTP not working
3843 => the returned list contains: plain, login, cram-md5, digest-md5, ...
3845 With the patch, sasl_listmech() also gets called three times. But now
3846 SASL's mech_list option is set to the server_mech specified in the the
3847 authenticator. Or in other words, the answer from sasl_listmech()
3848 gets limited to just the mech you're testing for (which is different
3850 => the return list contains just 'plain' or 'login', 'cram-md5' or
3851 nothing depending on the value of ob->server_mech.
3853 I've just tested the patch: Authentication still works fine,
3854 unavailable mechs specified in the exim configuration are still
3855 caught, and the auth.log warnings about OTP are gone."
3857 PH/25 When debugging is enabled, the contents of the command line are added
3858 to the debugging output, even when log_selector=+arguments is not
3861 PH/26 Change scripts/os-type so that when "uname -s" returns just "GNU", the
3862 answer is "GNU", and only if the return is "GNU/something" is the answer
3865 PH/27 $acl_verify_message is now set immediately after the failure of a
3866 verification in an ACL, and so is available in subsequent modifiers. In
3867 particular, the message can be preserved by coding like this:
3869 warn !verify = sender
3870 set acl_m0 = $acl_verify_message
3872 Previously, $acl_verify_message was set only while expanding "message"
3873 and "log_message" when a very denied access.
3875 PH/28 Modified OS/os.c-Linux with
3877 -#ifndef OS_LOAD_AVERAGE
3878 +#if !defined(OS_LOAD_AVERAGE) && defined(__linux__)
3880 to make Exim compile on kfreebsd-gnu. (I'm totally confused about the
3881 nomenclature these days.)
3883 PH/29 Installed patch from the Sieve maintainer that adds the options
3884 sieve_useraddress and sieve_subaddress to the redirect router.
3886 PH/30 In these circumstances:
3887 . Two addresses routed to the same list of hosts;
3888 . First host does not offer TLS;
3889 . First host accepts first address;
3890 . First host gives temporary error to second address;
3891 . Second host offers TLS and a TLS session is established;
3892 . Second host accepts second address.
3893 Exim incorrectly logged both deliveries with the TLS parameters (cipher
3894 and peerdn, if requested) that were in fact used only for the second
3897 PH/31 When doing a callout as part of verifying an address, Exim was not paying
3898 attention to any local part prefix or suffix that was matched by the
3899 router that accepted the address. It now behaves in the same way as it
3900 does for delivery: the affixes are removed from the local part unless
3901 rcpt_include_affixes is set on the transport.
3903 PH/32 Add the sender address, as F=<...>, to the log line when logging a
3904 timeout during the DATA phase of an incoming message.
3906 PH/33 Sieve envelope tests were broken for match types other than :is. I have
3907 applied a patch sanctioned by the Sieve maintainer.
3909 PH/34 Change 4.50/80 broke Exim in that it could no longer handle cases where
3910 the uid or gid is negative. A case of a negative gid caused this to be
3911 noticed. The fix allows for either to be negative.
3913 PH/35 ACL_WHERE_MIME is now declared unconditionally, to avoid too much code
3914 clutter, but the tables that are indexed by ACL_WHERE_xxx values had been
3917 PH/36 The change PH/12 above was broken. Fixed it.
3919 PH/37 Exim used to check for duplicate addresses in the middle of routing, on
3920 the grounds that routing the same address twice would always produce the
3921 same answer. This might have been true once, but it is certainly no
3922 longer true now. Routing a child address may depend on the previous
3923 routing that produced that child. Some complicated redirection strategies
3924 went wrong when messages had multiple recipients, and made Exim's
3925 behaviour dependent on the order in which the addresses were given.
3927 I have moved the duplicate checking until after the routing is complete.
3928 Exim scans the addresses that are assigned to local and remote
3929 transports, and removes any duplicates. This means that more work will be
3930 done, as duplicates will always all be routed, but duplicates are
3931 presumably rare, so I don't expect this is of any significance.
3933 For deliveries to pipes, files, and autoreplies, the duplicate checking
3934 still happens during the routing process, since they are not going to be
3937 PH/38 Installed a patch from Ian Freislich, with the agreement of Tom Kistner.
3938 It corrects a timeout issue with spamd. This is Ian's comment: "The
3939 background is that sometimes spamd either never reads data from a
3940 connection it has accepted, or it never writes response data. The exiscan
3941 spam.[ch] uses a 3600 second timeout on spamd socket reads, further, it
3942 blindly assumes that writes won't block so it may never time out."
3944 PH/39 Allow G after quota size as well as K and M.
3946 PH/40 The value set for $authenticated_id in an authenticator may not contain
3947 binary zeroes or newlines because the value is written to log lines and
3948 to spool files. There was no check on this. Now the value is run through
3949 the string_printing() function so that such characters are converted to
3950 printable escape sequences.
3952 PH/41 $message_linecount is a new variable that contains the total number of
3953 lines in the message. Compare $body_linecount, which is the count for the
3956 PH/42 Exim no longer gives details of delivery errors for specific addresses in
3957 bounce and delay warning messages, except in certain special cases, which
3960 (a) An SMTP error message from a remote host;
3961 (b) A message specified in a :fail: redirection;
3962 (c) A message specified in a "fail" command in a system filter;
3963 (d) A message specified in a FAIL return from the queryprogram router;
3964 (e) A message specified by the cannot_route_message router option.
3966 In these cases only, Exim does include the error details in bounce and
3967 warning messages. There are also a few cases where bland messages such
3968 as "unrouteable address" or "local delivery error" are given.
3970 PH/43 $value is now also set for the "else" part of a ${run expansion.
3972 PH/44 Applied patch from the Sieve maintainer: "The vacation draft is still
3973 being worked on, but at least Exim now implements the latest version to
3976 PH/45 In a pipe transport, although a timeout while waiting for the pipe
3977 process to complete was treated as a delivery failure, a timeout while
3978 writing the message to the pipe was logged, but erroneously treated as a
3979 successful delivery. Such timeouts include transport filter timeouts. For
3980 consistency with the overall process timeout, these timeouts are now
3981 treated as errors, giving rise to delivery failures by default. However,
3982 there is now a new Boolean option for the pipe transport called
3983 timeout_defer, which, if set TRUE, converts the failures into defers for
3984 both kinds of timeout. A transport filter timeout is now identified in
3987 PH/46 The "scripts/Configure-config.h" script calls "make" at one point. On
3988 systems where "make" and "gmake" are different, calling "gmake" at top
3989 level broke things. I've arranged for the value of $(MAKE) to be passed
3990 from the Makefile to this script so that it can call the same version of
3994 A note about Exim versions 4.44 and 4.50
3995 ----------------------------------------
3997 Exim 4.50 was meant to be the next release after 4.43. It contains a lot of
3998 changes of various kinds. As a consequence, a big documentation update was
3999 needed. This delayed the release for rather longer than seemed good, especially
4000 in the light of a couple of (minor) security issues. Therefore, the changes
4001 that fixed bugs were backported into 4.43, to create a 4.44 maintenance
4002 release. So 4.44 and 4.50 are in effect two different branches that both start
4005 I have left the 4.50 change log unchanged; it contains all the changes since
4006 4.43. The change log for 4.44 is below; many of its items are identical to
4007 those for 4.50. This seems to be the most sensible way to preserve the
4008 historical information.
4014 1. Minor wording change to the doc/README.SIEVE file.
4016 2. Change 4.43/35 introduced a bug: if quota_filecount was set, the
4017 computation of the current number of files was incorrect.
4019 3. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4020 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4023 4. Give more explanation in the error message when the command for a transport
4024 filter fails to execute.
4026 5. There are several places where Exim runs a non-Exim command in a
4027 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4028 was being done only for the command run by the queryprogram router. It is
4029 now done for all such subprocesses. The other cases are: ${run, transport
4030 filters, and the commands run by the lmtp and pipe transports.
4032 6. Added CONFIGURE_GROUP build-time option.
4034 7. Some older OS have a limit of 256 on the maximum number of file
4035 descriptors. Exim was using setrlimit() to set 1000 as a large value
4036 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4037 systems. I've change it so that if it can't get 1000, it tries for 256.
4039 8. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4040 was an oversight, and furthermore, ever since the addition of extra
4041 controls (e.g. 4.43/32), the checks on when to allow different forms of
4042 "control" were broken. There should now be diagnostics for all cases when a
4043 control that does not make sense is encountered.
4045 9. Added the /retain_sender option to "control=submission".
4047 10. $recipients is now available in the predata ACL (oversight).
4049 11. Tidy the search cache before the fork to do a delivery from a message
4050 received from the command line. Otherwise the child will trigger a lookup
4051 failure and thereby defer the delivery if it tries to use (for example) a
4052 cached ldap connection that the parent has called unbind on.
4054 12. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4055 of $address_data from the recipient verification was clobbered by the
4056 sender verification.
4058 13. The value of address_data from a sender verification is now available in
4059 $sender_address_data in subsequent conditions in the ACL statement.
4061 14. Added forbid_sieve_filter and forbid_exim_filter to the redirect router.
4063 15. Added a new option "connect=<time>" to callout options, to set a different
4066 16. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4067 was its contents. (It was OK if the option was not defined at all.)
4069 17. A "Completed" log line is now written for messages that are removed from
4070 the spool by the -Mrm option.
4072 18. New variables $sender_verify_failure and $recipient_verify_failure contain
4073 information about exactly what failed.
4075 19. Added -dd to debug only the daemon process.
4077 20. Incorporated Michael Haardt's patch to ldap.c for improving the way it
4078 handles timeouts, both on the server side and network timeouts. Renamed the
4079 CONNECT parameter as NETTIMEOUT (but kept the old name for compatibility).
4081 21. The rare case of EHLO->STARTTLS->HELO was setting the protocol to "smtp".
4082 It is now set to "smtps".
4084 22. $host_address is now set to the target address during the checking of
4085 ignore_target_hosts.
4087 23. When checking ignore_target_hosts for an ipliteral router, no host name was
4088 being passed; this would have caused $sender_host_name to have been used if
4089 matching the list had actually called for a host name (not very likely,
4090 since this list is usually IP addresses). A host name is now passed as
4093 24. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4094 code that specifies a non-restarting handler (typically sigaction() in
4095 modern systems) in an attempt to fix a rare and obscure crash bug.
4097 25. Narrowed the window for a race in the daemon that could cause it to ignore
4098 SIGCHLD signals. This is not a major problem, because they are used only to
4099 wake it up if nothing else does.
4101 26. A malformed maildirsize file could cause Exim to calculate negative values
4102 for the mailbox size or file count. Odd effects could occur as a result.
4103 The maildirsize information is now recalculated if the size or filecount
4106 27. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4107 support for a long time. Removed HAVE_SYS_VFS_H.
4109 28. Installed the latest version of exipick from John Jetmore.
4111 29. In an address list, if the pattern was not a regular expression, an empty
4112 subject address (from a bounce message) matched only if the pattern was an
4113 empty string. Non-empty patterns were not even tested. This was the wrong
4114 because it is perfectly reasonable to use an empty address as part of a
4115 database query. An empty address is now tested by patterns that are
4116 lookups. However, all the other forms of pattern expect the subject to
4117 contain a local part and a domain, and therefore, for them, an empty
4118 address still always fails if the pattern is not itself empty.
4120 30. Exim went into a mad DNS loop when attempting to do a callout where the
4121 host was specified on an smtp transport, and looking it up yielded more
4122 than one IP address.
4124 31. Re-factored the code for checking spool and log partition space into a
4125 function that finds that data and another that does the check. The former
4126 is then used to implement four new variables: $spool_space, $log_space,
4127 $spool_inodes, and $log_inodes.
4129 32. The RFC2047 encoding function was originally intended for short strings
4130 such as real names; it was not keeping to the 75-character limit for
4131 encoded words that the RFC imposes. It now respects the limit, and
4132 generates multiple encoded words if necessary. To be on the safe side, I
4133 have increased the buffer size for the ${rfc2047: expansion operator from
4136 33. It is now permitted to omit both strings after an "if" condition; if the
4137 condition is true, the result is "true". As before, when the second string
4138 is omitted, a false condition yields an empty string. This makes it less
4139 cumbersome to write custom ACL and router conditions.
4141 34. Failure to deliver a bounce message always caused it to be frozen, even if
4142 there was an errors_to setting on the router. The errors_to setting is now
4145 35. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4146 canonical form (fully expanded) before being placed in
4147 $sender_host_address.
4149 36. The table in the code that translates DNS record types into text (T_A to
4150 "A" for instance) was missing entries for NS and CNAME. It is just possible
4151 that this could have caused confusion if both these types were looked up
4152 for the same domain, because the text type is used as part of Exim's
4153 per-process caching. But the chance of anyone hitting this buglet seems
4156 37. The dnsdb lookup has been extended in a number of ways.
4158 (1) There is a new type, "zns", which walks up the domain tree until it
4159 finds some nameserver records. It should be used with care.
4161 (2) There is a new type, "mxh", which is like "mx" except that it returns
4162 just the host names, not the priorities.
4164 (3) It is now possible to give a list of domains (or IP addresses) to be
4165 looked up. The behaviour when one of the lookups defers can be
4166 controlled by a keyword.
4168 (4) It is now possible to specify the separator character for use when
4169 multiple records are returned.
4171 38. The dnslists ACL condition has been extended: it is now possible to supply
4172 a list of IP addresses and/or domains to be looked up in a particular DNS
4175 39. Added log_selector=+queue_time_overall.
4177 40. When running the queue in the test harness, wait just a tad after forking a
4178 delivery process, to get repeatability of debugging output.
4180 41. Include certificate and key file names in error message when GnuTLS fails
4181 to set them up, because the GnuTLS error message doesn't include the name
4182 of the failing file when there is a problem reading it.
4184 42. Allow both -bf and -bF in the same test run.
4186 43. Did the same fix as 41 above for OpenSSL, which had the same infelicity.
4188 44. The "Exiscan patch" is now merged into the mainline Exim source.
4190 45. Sometimes the final signoff response after QUIT could fail to get
4191 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4192 < 0 before doing a fflush(). This bug looks as though it goes back to the
4193 introduction of TLS in release 3.20, but "sometimes" must have been rare
4194 because the tests only now provoked it.
4196 46. Reset the locale to "C" after calling embedded Perl, in case it was changed
4197 (this can affect the format of dates).
4199 47. exim_tidydb, when checking for the continued existence of a message for
4200 which it has found a message-specific retry record, was not finding
4201 messages that were in split spool directories. Consequently, it was
4202 deleting retry records that should have stayed in existence.
4204 48. Steve fixed some bugs in eximstats.
4206 49. The SPA authentication driver was not abandoning authentication and moving
4207 on to the next authenticator when an expansion was forced to fail,
4208 contradicting the general specification for all authenticators. Instead it
4209 was generating a temporary error. It now behaves as specified.
4211 50. The default ordering of permitted cipher suites for GnuTLS was pessimal
4212 (the order specifies the preference for clients). The order is now AES256,
4213 AES128, 3DES, ARCFOUR128.
4215 51. Small patch to Sieve code - explicitly set From: when generating an
4218 52. Exim crashed if a remote delivery caused a very long error message to be
4219 recorded - for instance if somebody sent an entire SpamAssassin report back
4220 as a large number of 550 error lines. This bug was coincidentally fixed by
4221 increasing the size of one of Exim's internal buffers (big_buffer) that
4222 happened as part of the Exiscan merge. However, to be on the safe side, I
4223 have made the code more robust (and fixed the comments that describe what
4226 53. Now that there can be additional text after "Completed" in log lines (if
4227 the queue_time_overall log selector is set), a one-byte patch to exigrep
4228 was needed to allow it to recognize "Completed" as not the last thing in
4231 54. The LDAP lookup was not handling a return of LDAP_RES_SEARCH_REFERENCE. A
4232 patch that reportedly fixes this has been added. I am not expert enough to
4233 create a test for it. This is what the patch creator wrote:
4235 "I found a little strange behaviour of ldap code when working with
4236 Windows 2003 AD Domain, where users was placed in more than one
4237 Organization Units. When I tried to give exim partial DN, the exit code
4238 of ldap_search was unknown to exim because of LDAP_RES_SEARCH_REFERENCE.
4239 But simultaneously result of request was absolutely normal ldap result,
4240 so I produce this patch..."
4242 Later: it seems that not all versions of LDAP support LDAP_RES_SEARCH_
4243 REFERENCE, so I have modified the code to exclude the patch when that macro
4246 55. Some experimental protocols are using DNS PTR records for new purposes. The
4247 keys for these records are domain names, not reversed IP addresses. The
4248 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4249 leaves it alone. Component reversal etc. now happens only for IP addresses.
4252 56. Improve error message when ldap_search() fails in OpenLDAP or Solaris LDAP.
4254 57. Double the size of the debug message buffer (to 2048) so that more of very
4255 long debug lines gets shown.
4257 58. The exicyclog utility now does better if the number of log files to keep
4258 exceeds 99. In this case, it numbers them 001, 002 ... instead of 01, 02...
4260 59. Two changes related to the smtp_active_hostname option:
4262 (1) $smtp_active_hostname is now available as a variable.
4263 (2) The default for smtp_banner uses $smtp_active_hostname instead
4264 of $primary_hostname.
4266 60. The host_aton() function is supposed to be passed a string that is known
4267 to be a valid IP address. However, in the case of IPv6 addresses, it was
4268 not checking this. This is a hostage to fortune. Exim now panics and dies
4269 if the condition is not met. A case was found where this could be provoked
4270 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4271 components; fortuitously, this particular loophole had already been fixed
4272 by change 4.50/55 above.
4274 If there are any other similar loopholes, the new check in host_aton()
4275 itself should stop them being exploited. The report I received stated that
4276 data on the command line could provoke the exploit when Exim was running as
4277 exim, but did not say which command line option was involved. All I could
4278 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4279 running as the user.
4282 61. There was a buffer overflow vulnerability in the SPA authentication code
4283 (which came originally from the Samba project). I have added a test to the
4284 spa_base64_to_bits() function which I hope fixes it.
4287 62. Configuration update for GNU/Hurd and variations. Updated Makefile-GNU and
4288 os.h-GNU, and added configuration files for GNUkFreeBSD and GNUkNetBSD.
4290 63. The daemon start-up calls getloadavg() while still root for those OS that
4291 need the first call to be done as root, but it missed one case: when
4292 deliver_queue_load_max is set with deliver_drop_privilege. This is
4293 necessary for the benefit of the queue runner, because there is no re-exec
4294 when deliver_drop_privilege is set.
4296 64. A call to exiwhat cut short delays set up by "delay" modifiers in ACLs.
4297 This has been fixed.
4299 65. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4300 was in use, was not putting the data itself into the right store pool;
4301 consequently, it could be overwritten for a subsequent message in the same
4302 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4305 66. Added hosts_max_try_hardlimit to the smtp transport, default 50.
4307 67. The string_is_ip_address() function returns 0, 4, or 6, for "no an IP
4308 address", "IPv4 address", and "IPv6 address", respectively. Some calls of
4309 the function were treating the return as a boolean value, which happened to
4310 work because 0=false and not-0=true, but is not correct code.
4312 68. The host_aton() function was not handling scoped IPv6 addresses (those
4313 with, for example, "%eth0" on the end) correctly.
4315 69. Fixed some compiler warnings in acl.c for the bitmaps specified with
4316 negated items (that is, ~something) in unsigned ints. Some compilers
4317 apparently mutter when there is no cast.
4319 70. If an address verification called from an ACL failed, and did not produce a
4320 user-specific message (i.e. there was only a "system" message), nothing was
4321 put in $acl_verify_message. In this situation, it now puts the system
4324 71. Change 4.23/11 added synchronization checking at the start of an SMTP
4325 session; change 4.31/43 added the unwanted input to the log line - except
4326 that it did not do this in the start of session case. It now does.
4328 72. After a timeout in a callout SMTP session, Exim still sent a QUIT command.
4329 This is wrong and can cause the other end to generate a synchronization
4330 error if it is another Exim or anything else that does the synchronization
4331 check. A QUIT command is no longer sent after a timeout.
4333 73. $host_lookup_deferred has been added, to make it easier to detect DEFERs
4334 during host lookups.
4336 74. The defer_ok option of callout verification was not working if it was used
4337 when verifying addresses in header lines, that is, for this case:
4339 verify = header_sender/callout=defer_ok
4341 75. A backgrounded daemon closed stdin/stdout/stderr on entry; this meant that
4342 those file descriptors could be used for SMTP connections. If anything
4343 wrote to stderr (the example that came up was "warn" in embedded Perl), it
4344 could be sent to the SMTP client, causing chaos. The daemon now opens
4345 stdin, stdout, and stderr to /dev/null when it puts itself into the
4348 76. Arrange for output from Perl's "warn" command to be written to Exim's main
4349 log by default. The user can override this with suitable Perl magic.
4351 77. The use of log_message on a "discard" ACL verb, which is supposed to add to
4352 the log message when discard triggers, was not working for the DATA ACL or
4353 for the non-SMTP ACL.
4355 78. Error message wording change in sieve.c.
4357 79. If smtp_accept_max_per_host was set, the number of connections could be
4358 restricted to fewer than expected, because the daemon was trying to set up
4359 a new connection before checking whether the processes handling previous
4360 connections had finished. The check for completed processes is now done
4361 earlier. On busy systems, this bug wouldn't be noticed because something
4362 else would have woken the daemon, and it would have reaped the completed
4365 80. If a message was submitted locally by a user whose login name contained one
4366 or more spaces (ugh!), the spool file that Exim wrote was not re-readable.
4367 It caused a spool format error. I have fixed the spool reading code. A
4368 related problem was that the "from" clause in the Received: line became
4369 illegal because of the space(s). It is now covered by ${quote_local_part.
4371 81. Included the latest eximstats from Steve (adds average sizes to HTML Top
4374 82. Updated OS/Makefile-AIX as per message from Mike Meredith.
4376 83. Patch from Sieve maintainer to fix unterminated string problem in
4377 "vacation" handling.
4379 84. Some minor changes to the Linux configuration files to help with other
4380 OS variants using glibc.
4382 85. One more patch for Sieve to update vacation handling to latest spec.
4385 ----------------------------------------------------
4386 See the note above about the 4.44 and 4.50 releases.
4387 ----------------------------------------------------
4393 1. Change 4.43/35 introduced a bug that caused file counts to be
4394 incorrectly computed when quota_filecount was set in an appendfile
4397 2. Closing a stable door: arrange to panic-die if setitimer() ever fails. The
4398 bug fixed in 4.43/37 would have been diagnosed quickly if this had been in
4401 3. Give more explanation in the error message when the command for a transport
4402 filter fails to execute.
4404 4. There are several places where Exim runs a non-Exim command in a
4405 subprocess. The SIGUSR1 signal should be disabled for these processes. This
4406 was being done only for the command run by the queryprogram router. It is
4407 now done for all such subprocesses. The other cases are: ${run, transport
4408 filters, and the commands run by the lmtp and pipe transports.
4410 5. Some older OS have a limit of 256 on the maximum number of file
4411 descriptors. Exim was using setrlimit() to set 1000 as a large value
4412 unlikely to be exceeded. Change 4.43/17 caused a lot of logging on these
4413 systems. I've change it so that if it can't get 1000, it tries for 256.
4415 6. "control=submission" was allowed, but had no effect, in a DATA ACL. This
4416 was an oversight, and furthermore, ever since the addition of extra
4417 controls (e.g. 4.43/32), the checks on when to allow different forms of
4418 "control" were broken. There should now be diagnostics for all cases when a
4419 control that does not make sense is encountered.
4421 7. $recipients is now available in the predata ACL (oversight).
4423 8. Tidy the search cache before the fork to do a delivery from a message
4424 received from the command line. Otherwise the child will trigger a lookup
4425 failure and thereby defer the delivery if it tries to use (for example) a
4426 cached ldap connection that the parent has called unbind on.
4428 9. If verify=recipient was followed by verify=sender in a RCPT ACL, the value
4429 of $address_data from the recipient verification was clobbered by the
4430 sender verification.
4432 10. If FIXED_NEVER_USERS was defined, but empty, Exim was assuming the uid 0
4433 was its contents. (It was OK if the option was not defined at all.)
4435 11. A "Completed" log line is now written for messages that are removed from
4436 the spool by the -Mrm option.
4438 12. $host_address is now set to the target address during the checking of
4439 ignore_target_hosts.
4441 13. When checking ignore_target_hosts for an ipliteral router, no host name was
4442 being passed; this would have caused $sender_host_name to have been used if
4443 matching the list had actually called for a host name (not very likely,
4444 since this list is usually IP addresses). A host name is now passed as
4447 14. Changed the calls that set up the SIGCHLD handler in the daemon to use the
4448 code that specifies a non-restarting handler (typically sigaction() in
4449 modern systems) in an attempt to fix a rare and obscure crash bug.
4451 15. Narrowed the window for a race in the daemon that could cause it to ignore
4452 SIGCHLD signals. This is not a major problem, because they are used only to
4453 wake it up if nothing else does.
4455 16. A malformed maildirsize file could cause Exim to calculate negative values
4456 for the mailbox size or file count. Odd effects could occur as a result.
4457 The maildirsize information is now recalculated if the size or filecount
4460 17. Added HAVE_SYS_STATVFS_H to the os.h file for Linux, as it has had this
4461 support for a long time. Removed HAVE_SYS_VFS_H.
4463 18. Updated exipick to current release from John Jetmore.
4465 19. Allow an empty sender to be matched against a lookup in an address list.
4466 Previously the only cases considered were a regular expression, or an
4469 20. Exim went into a mad DNS lookup loop when doing a callout where the
4470 host was specified on the transport, if the DNS lookup yielded more than
4473 21. The RFC2047 encoding function was originally intended for short strings
4474 such as real names; it was not keeping to the 75-character limit for
4475 encoded words that the RFC imposes. It now respects the limit, and
4476 generates multiple encoded words if necessary. To be on the safe side, I
4477 have increased the buffer size for the ${rfc2047: expansion operator from
4480 22. Failure to deliver a bounce message always caused it to be frozen, even if
4481 there was an errors_to setting on the router. The errors_to setting is now
4484 23. If an IPv6 address is given for -bh or -bhc, it is now converted to the
4485 canonical form (fully expanded) before being placed in
4486 $sender_host_address.
4488 24. Updated eximstats to version 1.33
4490 25. Include certificate and key file names in error message when GnuTLS fails
4491 to set them up, because the GnuTLS error message doesn't include the name
4492 of the failing file when there is a problem reading it.
4494 26. Expand error message when OpenSSL has problems setting up cert/key files.
4497 27. Reset the locale to "C" after calling embedded Perl, in case it was changed
4498 (this can affect the format of dates).
4500 28. exim_tidydb, when checking for the continued existence of a message for
4501 which it has found a message-specific retry record, was not finding
4502 messages that were in split spool directories. Consequently, it was
4503 deleting retry records that should have stayed in existence.
4505 29. eximstats updated to version 1.35
4506 1.34 - allow eximstats to parse syslog lines as well as mainlog lines
4507 1.35 - bugfix such that pie charts by volume are generated correctly
4509 30. The SPA authentication driver was not abandoning authentication and moving
4510 on to the next authenticator when an expansion was forced to fail,
4511 contradicting the general specification for all authenticators. Instead it
4512 was generating a temporary error. It now behaves as specified.
4514 31. The default ordering of permitted cipher suites for GnuTLS was pessimal
4515 (the order specifies the preference for clients). The order is now AES256,
4516 AES128, 3DES, ARCFOUR128.
4518 31. Small patch to Sieve code - explicitly set From: when generating an
4521 32. Exim crashed if a remote delivery caused a very long error message to be
4522 recorded - for instance if somebody sent an entire SpamAssassin report back
4523 as a large number of 550 error lines. This bug was coincidentally fixed by
4524 increasing the size of one of Exim's internal buffers (big_buffer) that
4525 happened as part of the Exiscan merge. However, to be on the safe side, I
4526 have made the code more robust (and fixed the comments that describe what
4529 33. Some experimental protocols are using DNS PTR records for new purposes. The
4530 keys for these records are domain names, not reversed IP addresses. The
4531 dnsdb PTR lookup now tests whether its key is an IP address. If not, it
4532 leaves it alone. Component reversal etc. now happens only for IP addresses.
4535 34. The host_aton() function is supposed to be passed a string that is known
4536 to be a valid IP address. However, in the case of IPv6 addresses, it was
4537 not checking this. This is a hostage to fortune. Exim now panics and dies
4538 if the condition is not met. A case was found where this could be provoked
4539 from a dnsdb PTR lookup with an IPv6 address that had more than 8
4540 components; fortuitously, this particular loophole had already been fixed
4541 by change 4.50/55 or 4.44/33 above.
4543 If there are any other similar loopholes, the new check in host_aton()
4544 itself should stop them being exploited. The report I received stated that
4545 data on the command line could provoke the exploit when Exim was running as
4546 exim, but did not say which command line option was involved. All I could
4547 find was the use of -be with a bad dnsdb PTR lookup, and in that case it is
4548 running as the user.
4551 35. There was a buffer overflow vulnerability in the SPA authentication code
4552 (which came originally from the Samba project). I have added a test to the
4553 spa_base64_to_bits() function which I hope fixes it.
4556 36. The daemon start-up calls getloadavg() while still root for those OS that
4557 need the first call to be done as root, but it missed one case: when
4558 deliver_queue_load_max is set with deliver_drop_privilege. This is
4559 necessary for the benefit of the queue runner, because there is no re-exec
4560 when deliver_drop_privilege is set.
4562 37. Caching of lookup data for "hosts =" ACL conditions, when a named host list
4563 was in use, was not putting the data itself into the right store pool;
4564 consequently, it could be overwritten for a subsequent message in the same
4565 SMTP connection. (Fix 4.40/11 dealt with the non-cache case, but overlooked
4568 38. Sometimes the final signoff response after QUIT could fail to get
4569 transmitted in the non-TLS case. Testing !tls_active instead of tls_active
4570 < 0 before doing a fflush(). This bug looks as though it goes back to the
4571 introduction of TLS in release 3.20, but "sometimes" must have been rare
4572 because the tests only now provoked it.
4578 1. Fixed a longstanding but relatively impotent bug: a long time ago, before
4579 PIPELINING, the function smtp_write_command() used to return TRUE or FALSE.
4580 Now it returns an integer. A number of calls were still expecting a T/F
4581 return. Fortuitously, in all cases, the tests worked in OK situations,
4582 which is the norm. However, things would have gone wrong on any write
4583 failures on the smtp file descriptor. This function is used when sending
4584 messages over SMTP and also when doing verify callouts.
4586 2. When Exim is called to do synchronous delivery of a locally submitted
4587 message (the -odf or -odi options), it no longer closes stderr before doing
4590 3. Implemented the mua_wrapper option.
4592 4. Implemented mx_fail_domains and srv_fail_domains for the dnslookup router.
4594 5. Implemented the functions header_remove(), header_testname(),
4595 header_add_at_position(), and receive_remove_recipient(), and exported them
4598 6. If an ACL "warn" statement specified the addition of headers, Exim already
4599 inserted X-ACL-Warn: at the start if there was no header name. However, it
4600 was not making this test for the second and subsequent header lines if
4601 there were newlines in the string. This meant that an invalid header could
4602 be inserted if Exim was badly configured.
4604 7. Allow an ACL "warn" statement to add header lines at the start or after all
4605 the Received: headers, as well as at the end.
4607 8. Added the rcpt_4xx retry error code.
4609 9. Added postmaster_mailfrom=xxx to callout verification option.
4611 10. Added mailfrom=xxxx to the callout verification option, for verify=
4614 11. ${substr_1_:xxxx} and ${substr__3:xxxx} are now diagnosed as syntax errors
4615 (they previously behaved as ${substr_1_0:xxxx} and ${substr:_0_3:xxxx}).
4617 12. Inserted some casts to stop certain compilers warning when using pointer
4618 differences as field lengths or precisions in printf-type calls (mostly
4619 affecting debugging statements).
4621 13. Added optional readline() support for -be (dynamically loaded).
4623 14. Obscure bug fix: if a message error (e.g. 4xx to MAIL) happened within the
4624 same clock tick as a message's arrival, so that its received time was the
4625 same as the "first fail" time on the retry record, and that message
4626 remained on the queue past the ultimate address timeout, every queue runner
4627 would try a delivery (because it was past the ultimate address timeout) but
4628 after another failure, the ultimate address timeout, which should have then
4629 bounced the address, did not kick in. This was a "< instead of <=" error;
4630 in most cases the first failure would have been in the next clock tick
4631 after the received time, and all would be well.
4633 15. The special items beginning with @ in domain lists (e.g. @mx_any) were not
4634 being recognized when the domain list was tested by the match_domain
4635 condition in an expansion string.
4637 16. Added the ${str2b64: operator.
4639 17. Exim was always calling setrlimit() to set a large limit for the number of
4640 processes, without checking whether the existing limit was already
4641 adequate. (It did check for the limit on file descriptors.) Furthermore,
4642 errors from getrlimit() and setrlimit() were being ignored. Now they are
4643 logged to the main and panic logs, but Exim does carry on, to try to do its
4644 job under whatever limits there are.
4646 18. Imported PCRE 5.0.
4648 19. Trivial typo in log message " temporarily refused connection" (the leading
4651 20. If the log selector return_path_on_delivery was set and an address was
4652 redirected to /dev/null, the delivery process crashed because it assumed
4653 that a return path would always be set for a "successful" delivery. In this
4654 case, the whole delivery is bypassed as an optimization, and therefore no
4657 21. Internal re-arrangement: the function for sending a challenge and reading
4658 a response while authentication was assuming a zero-terminated challenge
4659 string. It's now changed to take a pointer and a length, to allow for
4660 binary data in such strings.
4662 22. Added the cyrus_sasl authenticator (code supplied by MBM).
4664 23. Exim was not respecting finduser_retries when seeking the login of the
4665 uid under which it was called; it was always trying 10 times. (The default
4666 setting of finduser_retries is zero.) Also, it was sleeping after the final
4667 failure, which is pointless.
4669 24. Implemented tls_on_connect_ports.
4671 25. Implemented acl_smtp_predata.
4673 26. If the domain in control=submission is set empty, Exim assumes that the
4674 authenticated id is a complete email address when it generates From: or
4675 Sender: header lines.
4677 27. Added "#define SOCKLEN_T int" to OS/os.h-SCO and OS/os.h-SCO_SV. Also added
4678 definitions to OS/Makefile-SCO and OS/Makefile-SCO_SV that put basename,
4679 chown and chgrp in /bin and hostname in /usr/bin.
4681 28. Exim was keeping the "process log" file open after each use, just as it
4682 does for the main log. This opens the possibility of it remaining open for
4683 long periods when the USR1 signal hits a daemon. Occasional processlog
4684 errors were reported, that could have been caused by this. Anyway, it seems
4685 much more sensible not to leave this file open at all, so that is what now
4688 29. The long-running daemon process does not normally write to the log once it
4689 has entered its main loop, and it closes the log before doing so. This is
4690 so that log files can straightforwardly be renamed and moved. However,
4691 there are a couple of unusual error situations where the daemon does write
4692 log entries, and I had neglected to close the log afterwards.
4694 30. The text of an SMTP error response that was received during a remote
4695 delivery was being truncated at 512 bytes. This is too short for some of
4696 the long messages that one sometimes sees. I've increased the limit to
4699 31. It is now possible to make retry rules that apply only when a message has a
4700 specific sender, in particular, an empty sender.
4702 32. Added "control = enforce_sync" and "control = no_enforce_sync". This makes
4703 it possible to be selective about when SMTP synchronization is enforced.
4705 33. Added "control = caseful_local_part" and "control = "caselower_local_part".
4707 32. Implemented hosts_connection_nolog.
4709 33. Added an ACL for QUIT.
4711 34. Setting "delay_warning=" to disable warnings was not working; it gave a
4714 35. Added mailbox_size and mailbox_filecount to appendfile.
4716 36. Added control = no_multiline_responses to ACLs.
4718 37. There was a bug in the logic of the code that waits for the clock to tick
4719 in the case where the clock went backwards by a substantial amount such
4720 that the microsecond fraction of "now" was more than the microsecond
4721 fraction of "then" (but the whole seconds number was less).
4723 38. Added support for the libradius Radius client library this is found on
4724 FreeBSD (previously only the radiusclient library was supported).
4730 1. When certain lookups returned multiple values in the form name=value, the
4731 quoting of the values was not always being done properly. Specifically:
4732 (a) If the value started with a double quote, but contained no whitespace,
4734 (b) If the value contained whitespace other than a space character (i.e.
4735 tabs or newlines or carriage returns) it was not quoted.
4736 This fix has been applied to the mysql and pgsql lookups by writing a
4737 separate quoting function and calling it from the lookup code. The fix
4738 should probably also be applied to nisplus, ibase and oracle lookups, but
4739 since I cannot test any of those, I have not disturbed their existing code.
4741 2. A hit in the callout cache for a specific address caused a log line with no
4742 reason for rejecting RCPT. Now it says "Previous (cached) callout
4743 verification failure".
4745 3. There was an off-by-one bug in the queryprogram router. An over-long
4746 return line was truncated at 256 instead of 255 characters, thereby
4747 overflowing its buffer with the terminating zero. As well as fixing this, I
4748 have increased the buffer size to 1024 (and made a note to document this).
4750 4. If an interrupt, such as the USR1 signal that is send by exiwhat, arrives
4751 when Exim is waiting for an SMTP response from a remote server, Exim
4752 restarts its select() call on the socket, thereby resetting its timeout.
4753 This is not a problem when such interrupts are rare. Somebody set up a cron
4754 job to run exiwhat every 2 minutes, which is less than the normal select()
4755 timeout (5 or 10 minutes). This meant that the select() timeout never
4756 kicked in because it was always reset. I have fixed this by comparing the
4757 time when an interrupt arrives with the time at the start of the first call
4758 to select(). If more time than the timeout has elapsed, the interrupt is
4759 treated as a timeout.
4761 5. Some internal re-factoring in preparation for the addition of Sieve
4762 extensions (by MH). In particular, the "personal" test is moved to a
4763 separate function, and given an option for scanning Cc: and Bcc: (which is
4764 not set for Exim filters).
4766 6. When Exim created an email address using the login of the caller as the
4767 local part (e.g. when creating a From: or Sender: header line), it was not
4768 quoting the local part when it contained special characters such as @.
4770 7. Installed new OpenBSD configuration files.
4772 8. Reworded some messages for syntax errors in "and" and "or" conditions to
4773 try to make them clearer.
4775 9. Callout options, other than the timeout value, were being ignored when
4776 verifying sender addresses in header lines. For example, when using
4778 verify = header_sender/callout=no_cache
4780 the cache was (incorrectly) being used.
4782 10. Added a missing instance of ${EXE} to the exim_install script; this affects
4783 only the Cygwin environment.
4785 11. When return_path_on_delivery was set as a log selector, if different remote
4786 addresses in the same message used different return paths and parallel
4787 remote delivery occurred, the wrong values would sometimes be logged.
4788 (Whenever a remote delivery process finished, the return path value from
4789 the most recently started remote delivery process was logged.)
4791 12. RFC 3848 specifies standard names for the "with" phrase in Received: header
4792 lines when AUTH and/or TLS are in use. This is the "received protocol"
4793 field. Exim used to use "asmtp" for authenticated SMTP, without any
4794 indication (in the protocol name) for TLS use. Now it follows the RFC and
4795 uses "esmtpa" if the connection is authenticated, "esmtps" if it is
4796 encrypted, and "esmtpsa" if it is both encrypted and authenticated. These
4797 names appear in log lines as well as in Received: header lines.
4799 13. Installed MH's patches for Sieve to add the "copy" and "vacation"
4800 extensions, and comparison tests, and to fix some bugs.
4802 14. Changes to the "personal" filter test:
4804 (1) The test was buggy in that it was just doing the equivalent of
4805 "contains" tests on header lines. For example, if a user's address was
4806 anne@some.where, the "personal" test would incorrectly be true for
4808 To: susanne@some.where
4810 This test is now done by extracting each address from the header in turn,
4811 and checking the entire address. Other tests that are part of "personal"
4812 are now done using regular expressions (for example, to check local parts
4813 of addresses in From: header lines).
4815 (2) The list of non-personal local parts in From: addresses has been
4816 extended to include "listserv", "majordomo", "*-request", and "owner-*",
4817 taken from the Sieve specification recommendations.
4819 (3) If the message contains any header line starting with "List-" it is
4820 treated as non-personal.
4822 (4) The test for "circular" in the Subject: header line has been removed
4823 because it now seems ill-conceived.
4825 15. Minor typos in src/EDITME comments corrected.
4827 16. Installed latest exipick from John Jetmore.
4829 17. If headers_add on a router specified a text string that was too long for
4830 string_sprintf() - that is, longer than 8192 bytes - Exim panicked. The use
4831 of string_sprintf() is now avoided.
4833 18. $message_body_size was not set (it was always zero) when running the DATA
4834 ACL and the local_scan() function.
4836 19. For the "mail" command in an Exim filter, no default was being set for
4837 the once_repeat time, causing a random time value to be used if "once" was
4838 specified. (If the value happened to be <= 0, no repeat happened.) The
4839 default is now 0s, meaning "never repeat". The "vacation" command was OK
4840 (its default is 7d). It's somewhat surprising nobody ever noticed this bug
4841 (I found it when inspecting the code).
4843 20. There is now an overall timeout for performing a callout verification. It
4844 defaults to 4 times the callout timeout, which applies to individual SMTP
4845 commands during the callout. The overall timeout applies when there is more
4846 than one host that can be tried. The timeout is checked before trying the
4847 next host. This prevents very long delays if there are a large number of
4848 hosts and all are timing out (e.g. when the network connections are timing
4849 out). The value of the overall timeout can be changed by specifying an
4850 additional sub-option for "callout", called "maxwait". For example:
4852 verify = sender/callout=5s,maxwait=20s
4854 21. Add O_APPEND to the open() call for maildirsize files (Exim already seeks
4855 to the end before writing, but this should make it even safer).
4857 22. Exim was forgetting that it had advertised PIPELINING for the second and
4858 subsequent messages on an SMTP connection. It was also not resetting its
4859 memory on STARTTLS and an internal HELO.
4861 23. When Exim logs an SMTP synchronization error within a session, it now
4862 records whether PIPELINING has been advertised or not.
4864 24. Added 3 instances of "(long int)" casts to time_t variables that were being
4865 formatted using %ld, because on OpenBSD (and perhaps others), time_t is int
4866 rather than long int.
4868 25. Installed the latest Cygwin configuration files from the Cygwin maintainer.
4870 26. Added the never_mail option to autoreply.
4876 1. A reorganization of the code in order to implement 4.40/8 caused a daemon
4877 crash if the getsockname() call failed; this can happen if a connection is
4878 closed very soon after it is established. The problem was simply in the
4879 order in which certain operations were done, causing Exim to try to write
4880 to the SMTP stream before it had set up the file descriptor. The bug has
4881 been fixed by making things happen in the correct order.
4887 1. If "drop" was used in a DATA ACL, the SMTP output buffer was not flushed
4888 before the connection was closed, thus losing the rejection response.
4890 2. Commented out the definition of SOCKLEN_T in os.h-SunOS5. It is needed for
4891 some early Solaris releases, but causes trouble in current releases where
4892 socklen_t is defined.
4894 3. When std{in,out,err} are closed, re-open them to /dev/null so that they
4897 4. Minor refactoring of os.c-Linux to avoid compiler warning when IPv6 is not
4900 5. Refactoring in expand.c to improve memory usage. Pre-allocate a block so
4901 that releasing the top of it at the end releases what was used for sub-
4902 expansions (unless the block got too big). However, discard this block if
4903 the first thing is a variable or header, so that we can use its block when
4904 it is dynamic (useful for very large $message_headers, for example).
4906 6. Lookups now cache *every* query, not just the most recent. A new, separate
4907 store pool is used for this. It can be recovered when all lookup caches are
4908 flushed. Lookups now release memory at the end of their result strings.
4909 This has involved some general refactoring of the lookup sources.
4911 7. Some code has been added to the store_xxx() functions to reduce the amount
4912 of flapping under certain conditions.
4914 8. log_incoming_interface used to affect only the <= reception log lines. Now
4915 it causes the local interface and port to be added to several more SMTP log
4916 lines, for example "SMTP connection from", and rejection lines.
4918 9. The Sieve author supplied some patches for the doc/README.SIEVE file.
4920 10. Added a conditional definition of _BSD_SOCKLEN_T to os.h-Darwin.
4922 11. If $host_data was set by virtue of a hosts lookup in an ACL, its value
4923 could be overwritten at the end of the current message (or the start of a
4924 new message if it was set in a HELO ACL). The value is now preserved for
4925 the duration of the SMTP connection.
4927 12. If a transport had a headers_rewrite setting, and a matching header line
4928 contained an unqualified address, that address was qualified, even if it
4929 did not match any rewriting rules. The underlying bug was that the values
4930 of the flags that permit the existence of unqualified sender and recipient
4931 addresses in header lines (set by {sender,recipient}_unqualified_hosts for
4932 non-local messages, and by -bnq for local messages) were not being
4933 preserved with the message after it was received.
4935 13. When Exim was logging an SMTP synchronization error, it could sometimes log
4936 "next input=" as part of the text comprising the host identity instead of
4937 the correct text. The code was using the same buffer for two different
4938 strings. However, depending on which order the printing function evaluated
4939 its arguments, the bug did not always show up. Under Linux, for example, my
4940 test suite worked just fine.
4942 14. Exigrep contained a use of Perl's "our" scoping after change 4.31/70. This
4943 doesn't work with some older versions of Perl. It has been changed to "my",
4944 which in any case is probably the better facility to use.
4946 15. A really picky compiler found some instances of statements for creating
4947 error messages that either had too many or two few arguments for the format
4950 16. The size of the buffer for calls to the DNS resolver has been increased
4951 from 1024 to 2048. A larger buffer is needed when performing PTR lookups
4952 for addresses that have a lot of PTR records. This alleviates a problem; it
4953 does not fully solve it.
4955 17. A dnsdb lookup for PTR records that receives more data than will fit in the
4956 buffer now truncates the list and logs the incident, which is the same
4957 action as happens when Exim is looking up a host name and its aliases.
4958 Previously in this situation something unpredictable would happen;
4959 sometimes it was "internal error: store_reset failed".
4961 18. If a server dropped the connection unexpectedly when an Exim client was
4962 using GnuTLS and trying to read a response, the client delivery process
4963 crashed while trying to generate an error log message.
4965 19. If a "warn" verb in an ACL added multiple headers to a message in a single
4966 string, for example:
4968 warn message = H1: something\nH2: something
4970 the text was added as a single header line from Exim's point of view
4971 though it ended up OK in the delivered message. However, searching for the
4972 second and subsequent header lines using $h_h2: did not work. This has been
4973 fixed. Similarly, if a system filter added multiple headers in this way,
4974 the routers could not see them.
4976 20. Expanded the error message when iplsearch is called with an invalid key to
4977 suggest using net-iplsearch in a host list.
4979 21. When running tests using -bh, any delays imposed by "delay" modifiers in
4980 ACLs are no longer actually imposed (and a message to that effect is
4983 22. If a "gecos" field in a passwd entry contained escaped characters, in
4984 particular, if it contained a \" sequence, Exim got it wrong when building
4985 a From: or a Sender: header from that name. A second bug also caused
4986 incorrect handling when an unquoted " was present following a character
4987 that needed quoting.
4989 23. "{crypt}" as a password encryption mechanism for a "crypteq" expansion item
4990 was not being matched caselessly.
4992 24. Arranged for all hyphens in the exim.8 source to be escaped with
4995 25. Change 16 of 4.32, which reversed 71 or 4.31 didn't quite do the job
4996 properly. Recipient callout cache records were still being keyed to include
4997 the sender, even when use_sender was set false. This led to far more
4998 callouts that were necessary. The sender is no longer included in the key
4999 when use_sender is false.
5001 26. Added "control = submission" modifier to ACLs.
5003 27. Added the ${base62d: operator to decode base 62 numbers.
5005 28. dnsdb lookups can now access SRV records.
5007 29. CONFIGURE_OWNER can be set at build time to define an alternative owner for
5008 the configuration file.
5010 30. The debug message "delivering xxxxxx-xxxxxx-xx" is now output in verbose
5011 (-v) mode. This makes the output for a verbose queue run more intelligible.
5013 31. Added a use_postmaster feature to recipient callouts.
5015 32. Added the $body_zerocount variable, containing the number of binary zero
5016 bytes in the message body.
5018 33. The time of last modification of the "new" subdirectory is now used as the
5019 "mailbox time last read" when there is a quota error for a maildir
5022 34. Added string comparison operators lt, lti, le, lei, gt, gti, ge, gei.
5024 35. Added +ignore_unknown as a special item in host lists.
5026 36. Code for decoding IPv6 addresses in host lists is now included, even if
5027 IPv6 support is not being compiled. This fixes a bug in which an IPv6
5028 address was recognized as an IP address, but was then not correctly decoded
5029 into binary, causing unexpected and incorrect effects when compared with
5036 1. Very minor rewording of debugging text in manualroute to say "list of
5037 hosts" instead of "hostlist".
5039 2. If verify=header_syntax was set, and a header line with an unqualified
5040 address (no domain) and a large number of spaces between the end of the
5041 name and the colon was received, the reception process suffered a buffer
5042 overflow, and (when I tested it) crashed. This was caused by some obsolete
5043 code that should have been removed. The fix is to remove it!
5045 3. When running in the test harness, delay a bit after writing a bounce
5046 message to get a bit more predictability in the log output.
5048 4. Added a call to search_tidyup() just before forking a reception process. In
5049 theory, someone could use a lookup in the expansion of smtp_accept_max_
5050 per_host which, without the tidyup, could leave open a database connection.
5052 5. Added the variables $recipient_data and $sender_data which get set from a
5053 lookup success in an ACL "recipients" or "senders" condition, or a router
5054 "senders" option, similar to $domain_data and $local_part_data.
5056 6. Moved the writing of debug_print from before to after the "senders" test
5059 7. Change 4.31/66 (moving the time when the Received: is generated) caused
5060 problems for message scanning, either using a data ACL, or using
5061 local_scan() because the Received: header was not generated till after they
5062 were called (in order to set the time as the time of reception completion).
5063 I have revised the way this works. The header is now generated after the
5064 body is received, but before the ACL or local_scan() are called. After they
5065 are run, the timestamp in the header is updated.
5071 1. Change 4.24/6 introduced a bug because the SIGALRM handler was disabled
5072 before starting a queue runner without re-exec. This happened only when
5073 deliver_drop_privilege was set or when the Exim user was set to root. The
5074 effect of the bug was that timeouts during subsequent deliveries caused
5075 crashes instead of being properly handled. The handler is now left at its
5076 default (and expected) setting.
5078 2. The other case in which a daemon avoids a re-exec is to deliver an incoming
5079 message, again when deliver_drop_privilege is set or Exim is run as root.
5080 The bug described in (1) was not present in this case, but the tidying up
5081 of the other signals was missing. I have made the two cases consistent.
5083 3. The ignore_target_hosts setting on a manualroute router was being ignored
5084 for hosts that were looked up using the /MX notation.
5086 4. Added /ignore=<ip list> feature to @mx_any, @mx_primary, and @mx_secondary
5089 5. Change 4.31/55 was buggy, and broke when there was a rewriting rule that
5090 operated on the sender address. After changing the $sender_address to <>
5091 for the sender address verify, Exim was re-instated it as the original
5092 (before rewriting) address, but remembering that it had rewritten it, so it
5093 wasn't rewriting it again. This bug also had the effect of breaking the
5094 sender address verification caching when the sender address was rewritten.
5096 6. The ignore_target_hosts option was being ignored by the ipliteral router.
5097 This has been changed so that if the ip literal address matches
5098 ignore_target_hosts, the router declines.
5100 7. Added expansion conditions match_domain, match_address, and match_local_
5101 part (NOT match_host).
5103 8. The placeholder for the Received: header didn't have a length field set.
5105 9. Added code to Exim itself and to exim_lock to test for a specific race
5106 condition that could lead to file corruption when using MBX delivery. The
5107 issue is with the lockfile that is created in /tmp. If this file is removed
5108 after a process has opened it but before that process has acquired a lock,
5109 there is the potential for a second process to recreate the file and also
5110 acquire a lock. This could lead to two Exim processes writing to the file
5111 at the same time. The added code performs the same test as UW imapd; it
5112 checks after acquiring the lock that its file descriptor still refers to
5113 the same named file.
5115 10. The buffer for building added header lines was of fixed size, 8192 bytes.
5116 It is now parameterized by HEADER_ADD_BUFFER_SIZE and this can be adjusted
5119 11. Added the smtp_active_hostname option. If used, this will typically be made
5120 to depend on the incoming interface address. Because $interface_address is
5121 not set up until the daemon has forked a reception process, error responses
5122 that can happen earlier (such as "too many connections") no longer contain
5125 12. If an expansion in a condition on a "warn" statement fails because a lookup
5126 defers, the "warn" statement is abandoned, and the next ACL statement is
5127 processed. Previously this caused the whole ACL to be aborted.
5129 13. Added the iplsearch lookup type.
5131 14. Added ident_timeout as a log selector.
5133 15. Added tls_certificate_verified as a log selector.
5135 16. Added a global option tls_require_ciphers (compare the smtp transport
5136 option of the same name). This controls incoming TLS connections.
5138 17. I finally figured out how to make tls_require_ciphers do a similar thing
5139 in GNUtls to what it does in OpenSSL, that is, set up an appropriate list
5140 before starting the TLS session.
5142 18. Tabs are now shown as \t in -bP output.
5144 19. If the log selector return_path_on_delivery was set, Exim crashed when
5145 bouncing a message because it had too many Received: header lines.
5147 20. If two routers both had headers_remove settings, and the first one included
5148 a superfluous trailing colon, the final name in the first list and the
5149 first name in the second list were incorrectly joined into one item (with a
5150 colon in the middle).
5156 1. Added -C and -D options to the exinext utility, mainly to make it easier
5157 to include in the automated testing, but these could be helpful when
5158 multiple configurations are in use.
5160 2. The exinext utility was not formatting the output nicely when there was
5161 an alternate port involved in the retry record key, nor when there was a
5162 message id as well (for retries that were specific to a specific message
5163 and a specific host). It was also confused by IPv6 addresses, because of
5164 the additional colons they contain. I have fixed the IPv4 problem, and
5165 patched it up to do a reasonable job for IPv6.
5167 3. When there is an error after a MAIL, RCPT, or DATA SMTP command during
5168 delivery, the log line now contains "pipelined" if PIPELINING was used.
5170 4. An SMTP transport process used to panic and die if the bind() call to set
5171 an explicit outgoing interface failed. This has been changed; it is now
5172 treated in the same way as a connect() failure.
5174 5. A reference to $sender_host_name in the part of a conditional expansion
5175 that was being skipped was still causing a DNS lookup. This no longer
5178 6. The def: expansion condition was not recognizing references to header lines
5179 that used bh_ and bheader_.
5181 7. Added the _cache feature to named lists.
5183 8. The code for checking quota_filecount in the appendfile transport was
5184 allowing one more file than it should have been.
5186 9. For compatibility with Sendmail, the command line option
5194 and sets the incoming protocol and host name (for trusted callers). The
5195 host name and its colon can be omitted when only the protocol is to be set.
5196 Note the Exim already has two private options, -pd and -ps, that refer to
5197 embedded Perl. It is therefore impossible to set a protocol value of "d" or
5198 "s", but I don't think that's a major issue.
5200 10. A number of refactoring changes to the code, none of which should affect
5203 (a) The number of logging options was getting close to filling up the
5204 32-bit word that was used as a bit map. I have split them into two classes:
5205 those that are passed in the argument to log_write(), and those that are
5206 only ever tested independently outside of that function. These are now in
5207 separate 32-bit words, so there is plenty of room for expansion again.
5208 There is no change in the user interface or the logging behaviour.
5210 (b) When building, for example, log lines, the code previously used a
5211 macro that called string_cat() twice, in order to add two strings. This is
5212 not really sufficiently general. Furthermore, there was one instance where
5213 it was actually wrong because one of the argument was used twice, and in
5214 one call a function was used. (As it happened, calling the function twice
5215 did not affect the overall behaviour.) The macro has been replaced by a
5216 function that can join an arbitrary number of extra strings onto a growing
5219 (c) The code for expansion conditions now uses a table and a binary chop
5220 instead of a serial search (which was left over from when there were very
5221 few conditions). Also, it now recognizes conditions like "pam" even when
5222 the relevant support is not compiled in: a suitably worded error message is
5223 given if an attempt is made to use such a condition.
5225 11. Added ${time_interval:xxxxx}.
5227 12. A bug was causing one of the ddress fields not to be passed back correctly
5228 from remote delivery subprocesses. The field in question was not being
5229 subsequently used, so this caused to problems in practice.
5231 13. Added new log selectors queue_time and deliver_time.
5233 14. Might have fixed a bug in maildirsizefile handling that threw up
5234 "unexpected character" debug warnings, and recalculated the data
5235 unnecessarily. In any case, I expanded the warning message to give more
5238 15. Added the message "Restricted characters in address" to the statements in
5239 the default ACL that block characters like @ and % in local parts.
5241 16. Change 71 for release 4.31 proved to be much less benign that I imagined.
5242 Three changes have been made:
5244 (a) There was a serious bug; a negative response to MAIL caused the whole
5245 recipient domain to be cached as invalid, thereby blocking all messages
5246 to all local parts at the same domain, from all senders. This bug has
5247 been fixed. The domain is no longer cached after a negative response to
5248 MAIL if the sender used is not empty.
5250 (b) The default behaviour of using MAIL FROM:<> for recipient callouts has
5253 (c) A new callout option, "use_sender" has been added for people who want
5254 the modified behaviour.
5260 1. Removed "EXTRALIBS=-lwrap" from OS/Makefile-Unixware7 on the advice of
5263 2. Removed "LIBS = -lresolv" from OS/Makefile-Darwin as it is not needed, and
5264 indeed breaks things for older releases.
5266 3. Added additional logging to the case where there is a problem reading data
5267 from a filter that is running in a subprocess using a pipe, in order to
5268 try to track down a specific problem.
5270 4. Testing facility fudge: when running in the test harness and attempting
5271 to connect to 10.x.x.x (expecting a connection timeout) I'm now sometimes
5272 getting "No route to host". Convert this to a timeout.
5274 5. Define ICONV_ARG2_TYPE as "char **" for Unixware7 to avoid compiler
5277 6. Some OS don't have socklen_t but use size_t instead. This affects the
5278 fifth argument of getsockopt() amongst other things. This is now
5279 configurable by a macro called SOCKLEN_T which defaults to socklen_t, but
5280 can be set for individual OS. I have set it for SunOS5, OSF1, and
5281 Unixware7. Current versions of SunOS5 (aka Solaris) do have socklen_t, but
5282 some earlier ones do not.
5284 7. Change 4.30/15 was not doing the test caselessly.
5286 8. The standard form for an IPv6 address literal was being rejected by address
5287 parsing in, for example, MAIL and RCPT commands. An example of this kind of
5288 address is [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5289 this, as well as the form without the "IPv6" on the front (but only when
5290 address literals are enabled, of course).
5292 9. Added some casts to avoid compiler warnings in OS/os.c-Linux.
5294 10. Exim crashed if a message with an empty sender address specified by -f
5295 encountered a router with an errors_to setting. This could be provoked only
5296 by a command such as
5300 where an empty string was supplied; "<>" did not hit this bug.
5302 11. Installed PCRE release 4.5.
5304 12. If EHLO/HELO was rejected by an ACL, the value of $sender_helo_name
5305 remained set. It is now erased.
5307 13. exiqgrep wasn't working on MacOS X because it didn't correctly compute
5308 times from message ids (which are base 36 rather than the normal 62).
5310 14. "Expected" SMTP protocol errors that can arise when PIPELINING is in use
5311 were being counted as actual protocol errors, and logged if the log
5312 selector +smtp_protocol_error was set. One cannot be perfect in this test,
5313 but now, if PIPELINING has been advertised, RCPT following a rejected MAIL,
5314 and DATA following a set of rejected RCPTs do not count as protocol errors.
5315 In other words, Exim assumes they were pipelined, though this may not
5316 actually be the case. Of course, in all cases the client gets an
5317 appropriate error code.
5319 15. If a lookup fails in an ACL condition, a message about the failure may
5320 be available; it is used if testing the ACL cannot continue, because most
5321 such messages specify what the cause of the deferral is. However, some
5322 messages (e.g. "MYSQL: no data found") do not cause a defer. There was bug
5323 that caused an old message to be retained and used if a later statement
5324 caused a defer, replacing the real cause of the deferral.
5326 16. If an IP address had so many PTR records that the DNS lookup buffer
5327 was not large enough to hold them, Exim could crash while trying to process
5328 the truncated data. It now detects and logs this case.
5330 17. Further to 4.21/58, another change has been made: if (and only if) the
5331 first line of a message (the first header line) ends with CRLF, a bare LF
5332 in a subsequent header line has a space inserted after it, so as not to
5333 terminate the header.
5335 18. Refactoring: tidied an ugly bit of code in appendfile that copied data
5336 unnecessarily, used atoi() instead of strtol(), and didn't check the
5337 termination when getting file sizes from file names by regex.
5339 19. Completely re-implemented the support for maildirsize files, in the light
5340 of a number of problems with the previous contributed implementation
5341 (4.30/29). In particular:
5343 . If the quota is zero, the maildirsize file is maintained, but no quota is
5346 . If the maildir directory does not exist, it is created before any attempt
5347 to write a maildirsize file.
5349 . The quota value in the file is just a cache; if the quota is changed in
5350 the transport, the new value overrides.
5352 . A regular expression is available for excluding directories from the
5355 20. The autoreply transport checks the characters in options that define the
5356 message's headers; it allows continued headers, but it was checking with
5357 isspace() after an embedded newline instead of explicitly looking for a
5360 21. If all the "regular" hosts to which an address was routed had passed their
5361 expiry times, and had not reached their retry times, the address was
5362 bounced, even if fallback hosts were defined. Now Exim should go on to try
5365 22. Increased buffer sizes in the callout code from 1024 to 4096 to match the
5366 equivalent code in the SMTP transport. Some hosts send humungous responses
5367 to HELO/EHLO, more than 1024 it seems.
5369 23. Refactoring: code in filter.c used (void *) for "any old type" but this
5370 gives compiler warnings in some environments. I've now done it "properly",
5373 24. The replacement for inet_ntoa() that is used with gcc on IRIX systems
5374 (because of problems with the built-in one) was declared to return uschar *
5375 instead of char *, causing compiler failure.
5377 25. Fixed a file descriptor leak when processing alias/forward files.
5379 26. Fixed a minor format string issue in dbfn.c.
5381 27. Typo in exim.c: ("dmbnz" for "dbmnz").
5383 28. If a filter file refered to $h_xxx or $message_headers, and the headers
5384 contained RFC 2047 "words", Exim's memory could, under certain conditions,
5387 29. When a sender address is verified, it is cached, to save repeating the test
5388 when there is more than one recipient in a message. However, when the
5389 verification involves a callout, it is possible for different callout
5390 options to be set for different recipients. It is too complicated to keep
5391 track of this in the cache, so now Exim always runs a verification when a
5392 callout is required, relying on the callout cache for the optimization.
5393 The overhead is duplication of the address routing, but this should not be
5396 30. Fixed a bug in callout caching. If a RCPT command caused the sender address
5397 to be verified with callout=postmaster, and the main callout worked but the
5398 postmaster check failed, the verification correctly failed. However, if a
5399 subsequent RCPT command asked for sender verification *without* the
5400 postmaster check, incorrect caching caused this verification also to fail,
5403 31. Exim caches DNS lookup failures so as to avoid multiple timeouts; however,
5404 it was not caching the DNS options (qualify_single, search_parents) that
5405 were used when the lookup failed. A subsequent lookup with different
5406 options therefore always gave the same answer, though there were cases
5407 where it should not have. (Example: a "domains = !$mx_any" option on a
5408 dnslookup router: the "domains" option is always processed without any
5409 widening, but the router might have qualify_single set.) Now Exim uses the
5410 cached value only when the same options are set.
5412 32. Added John Jetmore's "exipick" utility to the distribution.
5414 33. GnuTLS: When an attempt to start a TLS session fails for any reason other
5415 than a timeout (e.g. a certificate is required, and is not provided), an
5416 Exim server now closes the connection immediately. Previously it waited for
5417 the client to close - but if the client is SSL, it seems that they each
5418 wait for each other, leading to a delay before one of them times out.
5420 34: GnuTLS: Updated the code to use the new GnuTLS 1.0.0 API. I have not
5421 maintained 0.8.x compatibility because I don't think many are using it, and
5422 it is clearly obsolete.
5424 35. Added TLS support for CRLs: a tls_crl global option and one for the smtp
5427 36. OpenSSL: $tls_certificate_verified was being set to 1 even if the
5428 client certificate was expired. A simple patch fixes this, though I don't
5429 understand the full logic of why the verify callback is called multiple
5432 37. OpenSSL: a patch from Robert Roselius: "Enable client-bug workaround.
5433 Versions of OpenSSL as of 0.9.6d include a 'CBC countermeasure' feature,
5434 which causes problems with some clients (such as the Certicom SSL Plus
5435 library used by Eudora). This option, SSL_OP_DONT_INSERT_EMPTY_FRAGMENTS,
5436 disables the coutermeasure allowing Eudora to connect."
5438 38. Exim was not checking that a write() to a log file succeeded. This could
5439 lead to Bad Things if a log got too big, in particular if it hit a file
5440 size limit. Exim now panics and dies if it cannot write to a log file, just
5441 as it does if it cannot open a log file.
5443 39. Modified OS/Makefile-Linux so that it now contains
5445 CFLAGS=-O -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE
5447 The two -D definitions ensure that Exim is compiled with large file
5448 support, which makes it possible to handle log files that are bigger than
5451 40. Fixed a subtle caching bug: if (in an ACL or a set of routers, for
5452 instance) a domain was checked against a named list that involved a lookup,
5453 causing $domain_data to be set, then another domain was checked against the
5454 same list, then the first domain was re-checked, the value of $domain_data
5455 after the final check could be wrong. In particular, if the second check
5456 failed, it could be set empty. This bug probably also applied to
5459 41. The strip_trailing_dot option was not being applied to the address given
5460 with the -f command-line option.
5462 42. The code for reading a message's header from the spool was incrementing
5463 $received_count, but never initializing it. This meant that the value was
5464 incorrect (doubled) while delivering a message in the same process in which
5465 it was received. In the most common configuration of Exim, this never
5466 happens - a fresh exec is done - but it can happen when
5467 deliver_drop_privilege is set.
5469 43. When Exim logs an SMTP synchronization error - client data sent too soon -
5470 it now includes up to 150 characters of the unexpected data in the log
5473 44. The exim_dbmbuild utility uses fixed size buffers for reading input lines
5474 and building data strings. The size of both of these buffers was 10 000
5475 bytes - far larger than anybody would *ever* want, thought I. Needless to
5476 say, somebody hit the limit. I have increased the maximum line length to
5477 20 000 and the maximum data length of concatenated lines to 100 000. I have
5478 also fixed two bugs, because there was no checking on these buffers. Tsk,
5479 tsk. Now exim_dbmbuild gives a message and exits with an error code if a
5480 buffer is too small.
5482 45. The exim_dbmbuild utility did not support quoted keys, as Exim does in
5483 lsearch lookups. Now it does.
5485 46. When parsing a route_list item in a manualroute router, a fixed-length
5486 buffer was used for the list of hosts. I made this 1024 bytes long,
5487 thinking that nobody would ever have a list of hosts that long. Wrong.
5488 Somebody had a whole pile of complicated expansion conditions, and the
5489 string was silently truncated, leading to an expansion error. It turns out
5490 that it is easier to change to an unlimited length (owing to other changes
5491 that have happened since this code was originally written) than to build
5492 structure for giving a limitation error. The length of the item that
5493 expands into the list of hosts is now unlimited.
5495 47. The lsearch lookup could not handle data where the length of text line was
5496 more than 4095 characters. Such lines were truncated, leading to shortened
5497 data being returned. It should now handle lines of any length.
5499 48. Minor wording revision: "cannot test xxx in yyy ACL" becomes "cannot test
5500 xxx condition in yyy ACL" (e.g. "cannot test domains condition in DATA
5503 49. Cosmetic tidy to scripts like exicyclog that are generated by globally
5504 replacing strings such as BIN_DIRECTORY in a source file: the replacement
5505 no longer happens in comment lines. A list of replacements is now placed
5506 at the head of all of the source files, except those whose only change is
5507 to replace PERL_COMMAND in the very first #! line.
5509 50. Replaced the slow insertion sort in queue.c, for sorting the list of
5510 messages on the queue, with a bottom-up merge sort, using code contributed
5511 by Michael Haardt. This should make operations like -bp somewhat faster on
5512 large queues. It won't affect queue runners, except when queue_run_in_order
5515 51. Installed eximstats 1.31 in the distribution.
5517 52. Added support for SRV lookups to the dnslookup router.
5519 53. If an ACL referred to $message_body or $message_body_end, the value was not
5520 reset for any messages that followed in the same SMTP session.
5522 54. The store-handling optimization for building very long strings was not
5523 differentiating between the different store pools. I don't think this
5524 actually made any difference in practice, but I've tidied it.
5526 55. While running the routers to verify a sender address, $sender_address
5527 was still set to the sender address. This is wrong, because when routing to
5528 send a bounce to the sender, it would be empty. Therefore, I have changed
5529 it so that, while verifying a sender address, $sender_address is set to <>.
5530 (There is no change to what happens when verifying a recipient address.)
5532 56. After finding MX (or SRV) records, Exim was doing a DNS lookup for the
5533 target A or AAAA records (if not already returned) without resetting the
5534 qualify_single or search_parents options of the DNS resolver. These are
5535 inappropriate in this case because the targets of MX and SRV records must
5536 be FQDNs. A broken DNS record could cause trouble if it happened to have a
5537 target that, when qualified, matched something in the local domain. These
5538 two options are now turned off when doing these lookups.
5540 57. It seems that at least some releases of Reiserfs (which does not have the
5541 concept of a fixed number of inodes) returns zero and not -1 for the
5542 number of available inodes. This interacted badly with check_spool_inodes,
5543 which assumed that -1 was the "no such thing" setting. What I have done is
5544 to check that the total number of inodes is greater than zero before doing
5545 the test of how many are available.
5547 58. When a "warn" ACL statement has a log_message modifier, the message is
5548 remembered, and not repeated. This is to avoid a lot of repetition when a
5549 message has many recipients that cause the same warning to be written.
5550 Howewer, Exim was preserving the list of already written lines for an
5551 entire SMTP session, which doesn't seem right. The memory is now reset if a
5552 new message is started.
5554 59. The "rewrite" debugging flag was not showing the result of rewriting in the
5555 debugging output unless log_rewrite was also set.
5557 60. Avoid a compiler warning on 64-bit systems in dsearch.c by avoiding the use
5558 of (int)(handle) when we know that handle contains (void *)(-1).
5560 61. The Exim daemon panic-logs an error return when it closes the incoming
5561 connection. However "connection reset by peer" seems to be common, and
5562 isn't really an error worthy of noting specially, so that particular error
5565 62. When Exim is trying to find all the local interfaces, it used to panic and
5566 die if the ioctl to get the interface flags failed. However, it seems that
5567 on at least one OS (Solaris 9) it is possible to have an interface that is
5568 included in the list of interfaces, but for which you get a failure error
5569 for this call. This happens when the interface is not "plumbed" into a
5570 protocol (i.e. neither IPv4 nor IPv6). I've changed the code so that a
5571 failure of the "get flags" call assumes that the interface is down.
5573 63. Added a ${eval10: operator, which assumes all numbers are decimal. This
5574 makes life easier for people who are doing arithmetic on fields extracted
5575 from dates, where you often get leading zeros that should not be
5576 interpreted as octal.
5578 64. Added qualify_domain to the redirect router, to override the global
5581 65. If a pathologically long header line contained very many addresses (the
5582 report of this problem mentioned 10 000) and each of them was rewritten,
5583 Exim could use up a very large amount of memory. (It kept on making new
5584 copies of the header line as it rewrote, and never released the old ones.)
5585 At the expense of a bit more processing, the header rewriting function has
5586 been changed so that it no longer eats memory in this way.
5588 66. The generation of the Received: header has been moved from the time that a
5589 message starts to be received, to the time that it finishes. The timestamp
5590 in the Received: header should now be very close to that of the <= log
5591 line. There are two side-effects of this change:
5593 (a) If a message is rejected by a DATA or non-SMTP ACL or local_scan(), the
5594 logged header lines no longer include the local Received: line, because
5595 it has not yet been created. The same applies to a copy of the message
5596 that is returned to a non-SMTP sender when a message is rejected.
5598 (b) When a filter file is tested using -bf, no additional Received: header
5599 is added to the test message. After some thought, I decided that this
5602 This change does not affect the value of $received_for. It is still set
5603 after address rewriting, but before local_scan() is called.
5605 67. Installed the latest Cygwin-specific files from the Cygwin maintainer.
5607 68. GnuTLS: If an empty file is specified for tls_verify_certificates, GnuTLS
5608 gave an unhelpful panic error message, and a defer error. I have managed to
5609 change this behaviour so that it now rejects any supplied certificate,
5610 which seems right, as the list of acceptable certificates is empty.
5612 69. OpenSSL: If an empty file is specified for tls_verify_certificates, OpenSSL
5613 gave an unhelpful defer error. I have not managed to make this reject any
5614 supplied certificates, but the error message it gives is "no certificate
5615 supplied", which is not helpful.
5617 70. exigrep's output now also includes lines that are not associated with any
5618 message, but which match the given pattern. Implemented by a patch from
5619 Martin Sluka, which also tidied up the Perl a bit.
5621 71. Recipient callout verification, like sender verification, was using <> in
5622 the MAIL FROM command. This isn't really the right thing, since the actual
5623 sender may affect whether the remote host accepts the recipient or not. I
5624 have changed it to use the actual sender in the callout; this means that
5625 the cache record is now keyed on a recipient/sender pair, not just the
5626 recipient address. There doesn't seem to be a real danger of callout loops,
5627 since a callout by the remote host to check the sender would use <>.
5628 [SEE ABOVE: changed after hitting problems.]
5630 72. Exim treats illegal SMTP error codes that do not begin with 4 or 5 as
5631 temporary errors. However, in the case of such a code being given after
5632 the end of a data transmission (i.e. after ".") Exim was failing to write
5633 a retry record for the message. (Yes, there was some broken host that was
5634 actually sending 8xx at this point.)
5636 73. An unknown lookup type in a host list could cause Exim to panic-die when
5637 the list was checked. (An example that provoked this was putting <; in the
5638 middle of a list instead of at the start.) If this happened during a DATA
5639 ACL check, a -D file could be left lying around. This kind of configuration
5640 error no longer causes Exim to die; instead it causes a defer errror. The
5641 incident is still logged to the main and panic logs.
5643 74. Buglet left over from Exim 3 conversion. The message "too many messages
5644 in one connection" was written to the rejectlog but not the mainlog, except
5645 when address rewriting (yes!) was being logged.
5647 75. Added write_rejectlog option.
5649 76. When a system filter was run not as root (that is, when system_filter_user
5650 was set), the values of the $n variables were not being returned to the
5651 main process; thus, they were not subsequently available in the $sn
5654 77. Added +return_path_on_delivery log selector.
5656 78. A connection timeout was being treated differently from recipients deferred
5657 when testing hosts_max_try with a message that was older than the host's
5658 retry timeout. (The host should not be counted, thus allowing all hosts to
5659 be tried at least once before bouncing.) This may have been the cause of an
5660 occasionally reported bug whereby a message would remain on the queue
5661 longer than the retry timeout, but would be bounced if a delivery was
5662 forced. I say "may" because I never totally pinned down the problem;
5663 setting up timeout/retry tests is difficult. See also the next item.
5665 79. The ultimate address timeout was not being applied to errors that involved
5666 a combination of host plus message (for example, a timeout on a MAIL
5667 command). When an address resolved to a number of possible hosts, and they
5668 were not all tried for each delivery (e.g. because of hosts_max_try), a
5669 message could remain on the queue longer than the retry timeout.
5671 80. Sieve bug: "stop" inside "elsif" was broken. Applied a patch from Michael
5674 81. Fixed an obscure SMTP outgoing bug which required at least the following
5675 conditions: (a) there was another message waiting for the same server;
5676 (b) the server returned 5xx to all RCPT commands in the first message so
5677 that the message was not completed; (c) the server dropped the connection
5678 or gave a negative response to the RSET that Exim sends to abort the
5679 transaction. The observed case was a dropped connection after DATA that had
5680 been sent in pipelining mode. That is, the server had advertised PIPELINING
5681 but was not implementing it correctly. The effect of the bug was incorrect
5682 behaviour, such as trying another host, and this could lead to a crash.
5688 1. The 3rd arguments to getsockname(), getpeername(), and accept() in exim.c
5689 and daemon.c were passed as pointers to ints; they should have been
5690 pointers to socklen_t variables (which are typically unsigned ints).
5692 2. Some signed/unsigned type warnings in the os.c file for Linux have been
5695 3. Fixed a really odd bug that affected only the testing scheme; patching a
5696 certain fixed string in the binary changed the value of another string that
5697 happened to be identical to the end of the original first string.
5699 4. When gethostbyname() (or equivalent) is passed an IP address as a "host
5700 name", it returns that address as the IP address. On some operating
5701 systems (e.g. Solaris), it also passes back the IP address string as the
5702 "host name". However, on others (e.g. Linux), it passes back an empty
5703 string. Exim wasn't checking for this, and was changing the host name to an
5704 empty string, assuming it had been canonicized.
5706 5. Although rare, it is permitted to have more than one PTR record for a given
5707 IP address. I thought that gethostbyaddr() or getipnodebyaddr() always gave
5708 all the names associated with an address, because they do in Solaris.
5709 However, it seems that they do not in Linux for data that comes from the
5710 DNS. If an address in /etc/hosts has multiple names, they _are_ all given.
5711 I found this out when I moved to a new Linux workstation and tried to run
5712 the Exim test suite.
5714 To get round this problem I have changed the code so that it now does its
5715 own call to the DNS to look up PTR records when searching for a host name.
5716 If nothing can be found in the DNS, it tries gethostbyaddr(), so that
5717 addresses that are only in /etc/hosts are still found.
5719 This behaviour is, however, controlled by an option called host_lookup_
5720 order, which defaults to "bydns:byaddr". If people want to use the other
5721 order, or indeed, just use one or the other means of lookup, they can
5722 specify it in this variable.
5724 6. If a PTR record yields an empty name, Exim treats it as non-existent. In
5725 some operating systems, this comes back from gethostbyaddr() as an empty
5726 string, and this is what Exim used to test for. However, it seems that in
5727 other systems, "." is yielded. Exim now tests for this case too.
5729 7. The values of check_spool_space and check_log_space are now held internally
5730 as a number of kilobytes instead of an absolute number of bytes. If a
5731 numbers is specified without 'K' or 'M', it is rounded up to the nearest
5732 kilobyte. This means that much larger values can be stored.
5734 8. Exim monitor: an attempt to get the action menu when not actually pointing
5735 at a message produces an empty menu entitled "No message selected". This
5736 works on Solaris (OpenWindows). However, XFree86 does not like a menu with
5737 no entries in it ("Shell widget menu has zero width and/or height"). So I
5738 have added a single, blank menu entry in this case.
5740 9. Added ${quote_local_part.
5742 10. MIME decoding is now applied to the contents of Subject: header lines when
5745 11. Now that a reference to $sender_host_address automatically causes a reverse
5746 lookup to occur if necessary (4.13/18), there is no need to arrange for a
5747 host lookup before query-style lookups in lists that might use this
5748 variable. This has therefore been abolished, and the "net-" prefix is no
5749 longer necessary for query-style lookups.
5751 12. The Makefile for SCO_SV contained a setting of LDFLAGS. This appears to
5752 have been a typo for LFLAGS, so it has been changed.
5754 13. The install script calls Exim with "-C /dev/null" in order to find the
5755 version number. If ALT_CONFIG_PREFIX was set, this caused an error message
5756 to be output. Howeve, since Exim outputs its version number before the
5757 error, it didn't break the script. It just looked ugly. I fixed this by
5758 always allowing "-C /dev/null" if the caller is root.
5760 14. Ignore overlarge ACL variable number when reading spool file - insurance
5761 against a later release with more variables having written the file.
5763 15. The standard form for an IPv6 address literal was being rejected by EHLO.
5764 Example: [IPv6:2002:c1ed:8229:10:202:2dff:fe07:a42a]. Exim now accepts
5765 this, as well as the form without the "IPv6" on the front.
5767 16. Added CHOWN_COMMAND=/usr/sbin/chown and LIBS=-lresolv to the
5768 OS/Makefile-Darwin file.
5770 17. Fixed typo in lookups/ldap.c: D_LOOKUP should be D_lookup. This applied
5771 only to LDAP libraries that do not have LDAP_OPT_DEREF.
5773 18. After change 4.21/52, "%ld" was used to format the contents of the $inode
5774 variable. However, some OS use ints for inodes. I've added cast to long int
5775 to get rid of the compiler warning.
5777 19. I had forgotten to lock out "/../" in configuration file names when
5778 ALT_CONFIG_PREFIX was set.
5780 20. Routers used for verification do not need to specify transports. However,
5781 if such a router generated a host list, and callout was configured, Exim
5782 crashed, because it could not find a port number from the (non-existent)
5783 transport. It now assumes port 25 in this circumstance.
5785 21. Added the -t option to exigrep.
5787 22. If LOOKUP_LSEARCH is defined, all three linear search methods (lsearch,
5788 wildlsearch, nwildlsearch) are compiled. LOOKUP_WILDLSEARCH and LOOKUP_
5789 NWILDLSEARCH are now obsolete, but retained for compatibility. If either of
5790 them is set, LOOKUP_LSEARCH is forced.
5792 23. "exim -bV" now outputs a list of lookups that are included in the binary.
5794 24. Added sender and host information to the "rejected by local_scan()" log
5795 line; previously there was no indication of these.
5797 25. Added .include_if_exists.
5799 26. Change 3.952/11 added an explicit directory sync on top of a file sync for
5800 Linux. It turns out that not all file systems support this. Apparently some
5801 versions of NFS do not. (It's rare to put Exim's spool on NFS, but people
5802 do it.) To cope with this, the error EINVAL, which means that sync-ing is
5803 not supported on the file descriptor, is now ignored when Exim is trying to
5804 sync a directory. This applies only to Linux.
5806 27. Added -DBIND_8_COMPAT to the CLFAGS setting for Darwin.
5808 28. In Darwin (MacOS X), the PAM headers are in /usr/include/pam and not in
5809 /usr/include/security. There's now a flag in OS/os.h-Darwin to cope with
5812 29. Added support for maildirsize files from supplied patch (modified a bit).
5814 30. The use of :fail: followed by an empty string could lead Exim to respond to
5815 sender verification failures with (e.g.):
5817 550 Verification failed for <xxx>
5818 550 Sender verify failed
5820 where the first response line was missing the '-' that indicates it is not
5821 the final line of the response.
5823 31. The loop for finding the name of the user that called Exim had a hardwired
5824 limit of 10; it now uses the value of finduser_retries, which is used for
5825 all other user lookups.
5827 32. Added $received_count variable, available in data and not_smtp ACLs, and at
5830 33. Exim was neglecting to zero errno before one call of strtol() when
5831 expanding a string and expecting an integer value. On some systems this
5832 resulted in spurious "integer overflow" errors. Also, it was casting the
5833 result into an int without checking.
5835 34. Testing for a connection timeout using "timeout_connect" in the retry rules
5836 did not work. The code looks as if it has *never* worked, though it appears
5837 to have been documented since at least releast 1.62. I have made it work.
5839 35. The "timeout_DNS" error in retry rules, also documented since at least
5840 1.62, also never worked. As it isn't clear exactly what this means, and
5841 clearly it isn't a major issue, I have abolished the feature by treating it
5842 as "timeout", and writing a warning to the main and panic logs.
5844 36. The display of retry rules for -brt wasn't always showing the error code
5847 37. Added new error conditions to retry rules: timeout_A, timeout_MX,
5848 timeout_connect_A, timeout_connect_MX.
5850 38. Rewriting the envelope sender at SMTP time did not allow it to be rewritten
5851 to the empty sender.
5853 39. The daemon was not analysing the content of -oX till after it had closed
5854 stderr and disconnected from the controlling terminal. This meant that any
5855 syntax errors were only noted on the panic log, and the return code from
5856 the command was 0. By re-arranging the code a little, I've made the
5857 decoding happen first, so such errors now appear on stderr, and the return
5858 code is 1. However, the actual setting up of the sockets still happens in
5859 the disconnected process, so errors there are still only recorded on the
5862 40. A daemon listener on a wildcard IPv6 socket that also accepts IPv4
5863 connections (as happens on some IP stacks) was logged at start up time as
5864 just listening for IPv6. It now logs "IPv6 with IPv4". This differentiates
5865 it from "IPv6 and IPv4", which means that two separate sockets are being
5868 41. The debug output for gethostbyname2() or getipnodebyname() failures now
5869 says whether AF_INET or AF_INET6 was passed as an argument.
5871 42. Exiwhat output was messed up when time zones were included in log
5874 43. Exiwhat now gives more information about the daemon's listening ports,
5875 and whether -tls-on-connect was used.
5877 44. The "port" option of the smtp transport is now expanded.
5879 45. A "message" modifier in a "warn" statement in a non-message ACL was being
5880 silently ignored. Now an error message is written to the main and panic
5883 46. There's a new ACL modifier called "logwrite" which writes to a log file
5884 as soon as it is encountered.
5886 47. Added $local_user_uid and $local_user_gid at routing time.
5888 48. Exim crashed when trying to verify a sender address that was being
5891 49. Exim was recognizing only a space character after ".include". It now also
5892 recognizes a tab character.
5894 50. Fixed several bugs in the Perl script that creates the exim.8 man page by
5895 extracting the relevant information from the specification. The man page no
5896 longer contains scrambled data for the -d option, and I've added a section
5897 at the front about calling Exim under different names.
5899 51. Added "extra_headers" argument to the "mail" command in filter files.
5901 52. Redirecting mail to an unqualified address in a Sieve filter caused Exim to
5904 53. Installed eximstats 1.29.
5906 54. Added transport_filter_timeout as a generic transport option.
5908 55. Exim no longer adds an empty Bcc: header to messages that have no To: or
5909 Cc: header lines. This was required by RFC 822, but it not required by RFC
5912 56. Exim used to add From:, Date:, and Message-Id: header lines to any
5913 incoming messages that did not have them. Now it does so only if the
5914 message originates locally, that is, if there is no associated remote host
5915 address. When Resent- header lines are present, this applies to the Resent-
5916 lines rather than the non-Resent- lines.
5918 57. Drop incoming SMTP connection after too many syntax or protocol errors. The
5919 limit is controlled by smtp_max_synprot_errors, defaulting to 3.
5921 58. Messages for configuration errors now include the name of the main
5922 configuration file - useful now that there may be more than one file in a
5923 list (.included file names were always shown).
5925 59. Change 4.21/82 (run initgroups() when starting the daemon) causes problems
5926 for those rare installations that do not start the daemon as root or run it
5927 setuid root. I've cut out the call to initgroups() if the daemon is not
5930 60. The Exim user and group can now be bound into the binary as text strings
5931 that are looked up at the start of Exim's processing.
5933 61. Applied a small patch for the Interbase code, supplied by Ard Biesheuvel.
5935 62. Added $mailstore_basename variable.
5937 63. Installed patch to sieve.c from Michael Haardt.
5939 64. When Exim failed to open the panic log after failing to open the main log,
5940 the original message it was trying to log was written to stderr and debug
5941 output, but if they were not available (the usual case in production), it
5942 was lost. Now it is written to syslog before the two lines that record the
5943 failures to open the logs.
5945 65. Users' Exim filters run in subprocesses under the user's uid. It is
5946 possible for a "deliver" command or an alias in a "personal" command to
5947 provoke an address rewrite. If logging of address rewriting is configured,
5948 this fails because the process is not running as root or exim. There may be
5949 a better way of dealing with this, but for the moment (because 4.30 needs
5950 to be released), I have disabled address rewrite logging when running a
5951 filter in a non-root, non-exim process.
5957 1. The buildconfig auxiliary program wasn't quoting the value set for
5958 HEADERS_CHARSET. This caused a compilation error complaining that 'ISO' was
5959 not defined. This bug was masked in 4.22 by the effect that was fixed in
5962 2. Some messages that were rejected after a message id was allocated were
5963 shown as "incomplete" by exigrep. It no longer does this for messages that
5964 are rejected by local_scan() or the DATA or non-SMTP ACLs.
5966 3. If a Message-ID: header used a domain literal in the ID, and Exim did not
5967 have allow_domain_literals set, the ID did not get logged in the <= line.
5968 Domain literals are now always recognized in Message-ID: header lines.
5970 4. The first argument for a ${extract expansion item is the key name or field
5971 number. Leading and trailing spaces in this item were not being ignored,
5972 causing some misleading effects.
5974 5. When deliver_drop_privilege was set, single queue runner processes started
5975 manually (i.e. by the command "exim -q") or by the daemon (which uses the
5976 same command in the process it spins off) were not dropping privilege.
5978 6. When the daemon running as "exim" started a queue runner, it always
5979 re-executed Exim in the spun-off process. This is a waste of effort when
5980 deliver_drop_privilege is set. The new process now just calls the
5981 queue-runner function directly.
5987 1. Typo in the src/EDITME file: it referred to HEADERS_DECODE_TO instead of
5990 2. Change 4.21/73 introduced a bug. The pid file path set by -oP was being
5991 ignored. Though the use of -oP was forcing the writing of a pid file, it
5992 was always written to the default place.
5994 3. If the message "no IP address found for host xxxx" is generated during
5995 incoming verification, it is now followed by identification of the incoming
5996 connection (so you can more easily find what provoked it).
5998 4. Bug fix for Sieve filters: "stop" inside a block was not working properly.
6000 5. Added some features to "harden" Exim a bit more against certain attacks:
6002 (a) There is now a build-time option called FIXED_NEVER_USERS that can
6003 be put in Local/Makefile. This is like the never_users runtime option,
6004 but it cannot be overridden. The default setting is "root".
6006 (b) If ALT_CONFIG_PREFIX is defined in Local/Makefile, it specifies a
6007 prefix string with which any file named in a -C command line option
6010 (c) If ALT_CONFIG_ROOT_ONLY is defined in Local/Makefile, root privilege
6011 is retained for -C and -D only if the caller of Exim is root. Without
6012 it, the exim user may also use -C and -D and retain privilege.
6014 (d) If DISABLE_D_OPTION is defined in Local/Makefile, the use of the -D
6015 command line option is disabled.
6017 6. Macro names set by the -D option must start with an upper case letter, just
6018 like macro names defined in the configuration file.
6020 7. Added "dereference=" facility to LDAP.
6022 8. Two instances of the typo "uknown" in the source files are fixed.
6024 9. If a PERL_COMMAND setting in Local/Makefile was not at the start of a line,
6025 the Configure-Makefile script screwed up while processing it.
6027 10. Incorporated PCRE 4.4.
6029 11. The SMTP synchronization check was not operating right at the start of an
6030 SMTP session. For example, it could not catch a HELO sent before the client
6031 waited for the greeting. There is now a check for outstanding input at the
6032 point when the greeting is written. Because of the duplex, asynchronous
6033 nature of TCP/IP, it cannot be perfect - the incorrect input may be on its
6034 way, but not yet received, when the check is performed.
6036 12. Added tcp_nodelay to make it possible to turn of the setting of TCP_NODELAY
6037 on TCP/IP sockets, because this apparently causes some broken clients to
6040 13. Installed revised OS/Makefile-CYGWIN and OS/os.c-cygwin (the .h file was
6041 unchanged) from the Cygwin maintainer.
6043 14. The code for -bV that shows what is in the binary showed "mbx" when maildir
6044 was supported instead of testing for mbx. Effectively a typo.
6046 15. The spa authenticator server code was not checking that the input it
6047 received was valid base64.
6049 16. The debug output line for the "set" modifier in ACLs was not showing the
6050 name of the variable that was being set.
6052 17. Code tidy: the variable type "vtype_string" was never used. Removed it.
6054 18. Previously, a reference to $sender_host_name did not cause a DNS reverse
6055 lookup on its own. Something else was needed to trigger the lookup. For
6056 example, a match in host_lookup or the need for a host name in a host list.
6057 Now, if $sender_host_name is referenced and the host name has not yet been
6058 looked up, a lookup is performed. If the lookup fails, the variable remains
6059 empty, and $host_lookup_failed is set to "1".
6061 19. Added "eqi" as a case-independent comparison operator.
6063 20. The saslauthd authentication condition could segfault if neither service
6064 nor realm was specified.
6066 21. If an overflowing value such as "2048M" was set for message_size_limit, the
6067 error message that was logged was misleading, and incoming SMTP
6068 connections were dropped. The message is now more accurate, and temporary
6069 errors are given to SMTP connections.
6071 22. In some error situations (such as 21 above) Exim rejects all SMTP commands
6072 (except RSET) with a 421 error, until QUIT is received. However, it was
6073 failing to send a response to QUIT.
6075 23. The HELO ACL was being run before the code for helo_try_verify_hosts,
6076 which made it impossible to use "verify = helo" in the HELO ACL. The HELO
6077 ACL is now run after the helo_try_verify_hosts code.
6079 24. "{MD5}" and "{SHA1}" are now recognized as equivalent to "{md5"} and
6080 "{sha1}" in the "crypteq" expansion condition (in fact the comparison is
6081 case-independent, so other case variants are also recognized). Apparently
6082 some systems use these upper case variants.
6084 25. If more than two messages were waiting for the same host, and a transport
6085 filter was specified for the transport, Exim sent two messages over the
6086 same TCP/IP connection, and then failed with "socket operation on non-
6087 socket" when it tried to send the third.
6089 26. Added Exim::debug_write and Exim::log_write for embedded Perl use.
6091 27. The extern definition of crypt16() in expand.c was not being excluded when
6092 the OS had its own crypt16() function.
6094 28. Added bounce_return_body as a new option, and bounce_return_size_limit
6095 as a preferred synonym for return_size_limit, both as an option and as an
6098 29. Added LIBS=-liconv to OS/Makefile-OSF1.
6100 30. Changed the default configuration ACL to relax the local part checking rule
6101 for addresses that are not in any local domains. For these addresses,
6102 slashes and pipe symbols are allowed within local parts, but the sequence
6103 /../ is explicitly forbidden.
6105 31. SPA server authentication was not clearing the challenge buffer before
6108 32. log_message in a "warn" ACL statement was writing to the reject log as
6109 well as to the main log, which contradicts the documentation and doesn't
6110 seem right (because no rejection is happening). So I have stopped it.
6112 33. Added Ard Biesheuvel's lookup code for accessing an Interbase database.
6113 However, I am unable to do any testing of this.
6115 34. Fixed an infelicity in the appendfile transport. When checking directories
6116 for a mailbox, to see if any needed to be created, it was accidentally
6117 using path names with one or more superfluous leading slashes; tracing
6118 would show up entries such as stat("///home/ph10", 0xFFBEEA48).
6120 35. If log_message is set on a "discard" verb in a MAIL or RCPT ACL, its
6121 contents are added to the log line that is written for every discarded
6122 recipient. (Previously a log_message setting was ignored.)
6124 36. The ${quote: operator now quotes the string if it is empty.
6126 37. The install script runs exim in order to find its version number. If for
6127 some reason other than non-existence or emptiness, which it checks, it
6128 could not run './exim', it was installing it with an empty version number,
6129 i.e. as "exim-". This error state is now caught, and the installation is
6132 38. An argument was missing from the function that creates an error message
6133 when Exim fails to connect to the socket for saslauthd authentication.
6134 This could cause Exim to crash, or give a corrupted message.
6136 39. Added isip, isip4, and isip6 to ${if conditions.
6138 40. The ACL variables $acl_xx are now saved with the message, and can be
6139 accessed later in routers, transports, and filters.
6141 41. The new lookup type nwildlsearch is like wildlsearch, except that the key
6142 strings in the file are not string-expanded.
6144 42. If a MAIL command specified a SIZE value that was too large to fit into an
6145 int variable, the check against message_size_limit failed. Such values are
6146 now forced to INT_MAX, which is around 2Gb for a 32-bit variable. Maybe one
6147 day this will have to be increased, but I don't think I want to be around
6148 when emails are that large.
6155 1. Removed HAVE_ICONV=yes from OS/Makefile-FreeBSD, since it seems that
6156 iconv() is not standard in FreeBSD.
6158 2. Change 4.21/17 was buggy and could cause stack overwriting on a system with
6159 IPv6 enabled. The observed symptom was a segmentation fault on return from
6160 the function os_common_find_running_interfaces() in src/os.c.
6162 3. In the check_special_case() function in daemon.c I had used "errno" as an
6163 argument name, which causes warnings on some systems. This was basically a
6164 typo, since it was named "eno" in the comments!
6166 4. The code that waits for the clock to tick (at a resolution of some fraction
6167 of a second) so as to ensure message-id uniqueness was always waiting for
6168 at least one whole tick, when it could have waited for less. [This is
6169 almost certainly not relevant at current processor speeds, where it is
6170 unlikely to ever wait at all. But we try to future-proof.]
6172 5. The function that sleeps for a time interval that includes fractions of a
6173 second contained a race. It did not block SIGALRM between setting the
6174 timer, and suspending (a couple of lines later). If the interval was short
6175 and the sigsuspend() was delayed until after it had expired, the suspension
6176 never ended. On busy systems this could lead to processes getting stuck for
6179 6. Some uncommon configurations may cause a lookup to happen in a queue runner
6180 process, before it forks any delivery processes. The open lookup caching
6181 mechanism meant that the open file or database connection was passed into
6182 the delivery process. The problem was that delivery processes always tidy
6183 up cached lookup data. This could cause a problem for the next delivery
6184 process started by the queue runner, because the external queue runner
6185 process does not know about the closure. So the next delivery process
6186 still has data in the lookup cache. In the case of a file lookup, there was
6187 no problem because closing a file descriptor in a subprocess doesn't affect
6188 the parent. However, if the lookup was caching a connection to a database,
6189 the connection was closed, and the second delivery process was likely to
6190 see errors such as "PGSQL: query failed: server closed the connection
6191 unexpectedly". The problem has been fixed by closing all cached lookups
6192 in a queue runner before running a delivery process.
6194 7. Compiler warning on Linux for the second argument of iconv(), which doesn't
6195 seem to have the "const" qualifier which it has on other OS. I've
6198 8. Change 4.21/2 was too strict. It is only if there are two authenticators
6199 *of the same type* (client or server) with the same public name that an
6200 error should be diagnosed.
6202 9. When Exim looked up a host name for an IP address, but failed to find the
6203 original IP address when looking up the host name (a safety check), it
6204 output the message "<ip address> does not match any IP for NULL", which was
6205 confusing, to say the least. The bug was that the host name should have
6206 appeared instead of "NULL".
6208 10. Since release 3.03, if Exim is called by a uid other than root or the Exim
6209 user that is built into the binary, and the -C or -D options is used, root
6210 privilege is dropped before the configuration file is read. In addition,
6211 logging is switched to stderr instead of the normal log files. If the
6212 configuration then re-defines the Exim user, the unprivileged environment
6213 is probably not what is expected, so Exim logs a panic warning message (but
6216 However, if deliver_drop_privilege is set, the unprivileged state may well
6217 be exactly what is intended, so the warning has been cut out in that case,
6218 and Exim is allowed to try to write to its normal log files.
6224 1. smtp_return_error_details was not giving details for temporary sender
6225 or receiver verification errors.
6227 2. Diagnose a configuration error if two authenticators have the same public
6230 3. Exim used not to create the message log file for a message until the first
6231 delivery attempt. This could be confusing when incoming messages were held
6232 for policy or load reasons. The message log file is now created at the time
6233 the message is received, and an initial "Received" line is written to it.
6235 4. The automatically generated man page for command line options had a minor
6236 bug that caused no ill effects; however, a more serious problem was that
6237 the procedure for building the man page automatically didn't always
6238 operate. Consequently, release 4.20 contains an out-of-date version. This
6239 shouldn't happen again.
6241 5. When building Exim with embedded Perl support, the script that builds the
6242 Makefile was calling 'perl' to find its compile-time parameters, ignoring
6243 any setting of PERL_COMMAND in Local/Makefile. This is now fixed.
6245 6. The freeze_tell option was not being used for messages that were frozen on
6246 arrival, either by an ACL or by local_scan().
6248 7. Added the smtp_incomplete_transaction log selector.
6250 8. After STARTTLS, Exim was not forgetting that it had advertised AUTH, so it
6251 was accepting AUTH without a new EHLO.
6253 9. Added tls_remember_esmtp to cope with YAEB. This allows AUTH and other
6254 ESMTP extensions after STARTTLS without a new EHLO, in contravention of the
6257 10. Logging of TCP/IP connections (when configured) now happens in the main
6258 daemon process instead of the child process, so that the TCP/IP connection
6259 count is more accurate (but it can never be perfect).
6261 11. The use of "drop" in a nested ACL was not being handled correctly in the
6262 outer ACL. Now, if condition failure induced by the nested "drop" causes
6263 the outer ACL verb to deny access ("accept" or "discard" after "endpass",
6264 or "require"), the connection is dropped.
6266 12. Similarly, "discard" in a nested ACL wasn't being handled. A nested ACL
6267 that yield "discard" can now be used with an "accept" or a "discard" verb,
6268 but an error is generated for any others (because I can't see a useful way
6269 to define what should happen).
6271 13. When an ACL is read dynamically from a file (or anywhere else), the lines
6272 are now processed in the same way as lines in the Exim configuration file.
6273 In particular, continuation lines are supported.
6275 14. Added the "dnslists = a.b.c!=n.n.n.n" feature.
6277 15. Added -ti meaning -t -i.
6279 16. Check for letters, digits, hyphens, and dots in the names of dnslist
6280 domains, and warn by logging if others are found.
6282 17. At least on BSD, alignment is not guarenteed for the array of ifreq's
6283 returned from GIFCONF when Exim is trying to find the list of interfaces on
6284 a host. The code in os.c has been modified to copy each ifreq to an aligned
6285 structure in all cases.
6287 Also, in some cases, the returned ifreq's were being copied to a 'struct
6288 ifreq' on the stack, which was subsequently passed to host_ntoa(). That
6289 means the last couple of bytes of an IPv6 address could be chopped if the
6290 ifreq contained only a normal sockaddr (14 bytes storage).
6292 18. Named domain lists were not supported in the hosts_treat_as_local option.
6293 An entry such as +xxxx was not recognized, and was treated as a literal
6296 19. Ensure that header lines added by a DATA ACL are included in the reject log
6297 if the ACL subsequently rejects the message.
6299 20. Upgrade the cramtest.pl utility script to use Digest::MD5 instead of just
6300 MD5 (which is deprecated).
6302 21. When testing a filter file using -bf, Exim was writing a message when it
6303 took the sender from a "From " line in the message, but it was not doing so
6304 when it took $return_path from a Return-Path: header line. It now does.
6306 22. If the contents of a "message" modifier for a "warn" ACL verb do not begin
6307 with a valid header line field name (a series of printing characters
6308 terminated by a colon, Exim now inserts X-ACL-Warn: at the beginning.
6310 23. Changed "disc" in the source to "disk" to conform to the documentation and
6311 the book and for uniformity.
6313 24. Ignore Sendmail's -Ooption=value command line item.
6315 25. When execve() failed while trying to run a command in a pipe transport,
6316 Exim was returning EX_UNAVAILBLE (69) from the subprocess. However, this
6317 could be confused with a return value of 69 from the command itself. This
6318 has been changed to 127, the value the shell returns if it is asked to run
6319 a non-existent command. The wording for the related log line suggests a
6320 non-existent command as the problem.
6322 26. If received_header_text expands to an empty string, do not add a Received:
6323 header line to the message. (Well, it adds a token one on the spool, but
6324 marks it "old" so that it doesn't get used or transmitted.)
6326 27. Installed eximstats 1.28 (addition of -nt option).
6328 28. There was no check for failure on the call to getsockname() in the daemon
6329 code. This can fail if there is a shortage of resources on the system, with
6330 ENOMEM, for example. A temporary error is now given on failure.
6332 29. Contrary to the C standard, it seems that in some environments, the
6333 equivalent of setlocale(LC_ALL, "C") is not obeyed at the start of a C
6334 program. Exim now does this explicitly; it affects the formatting of
6335 timestamps using strftime().
6337 30. If exiqsumm was given junk data, it threw up some uninitialized variable
6338 complaints. I've now initialized all the variables, to avoid this.
6340 32. Header lines added by a system filter were not being "seen" during
6341 transport-time rewrites.
6343 33. The info_callback() function passed to OpenSSL is set up with type void
6344 (*)(SSL *, int, int), as described somewhere. However, when calling the
6345 function (actually a macro) that sets it up, the type void(*)() is
6346 expected. I've put in a cast to prevent warnings from picky compilers.
6348 34. If a DNS black list lookup found a CNAME record, but there were no A
6349 records associated with the domain it pointed at, Exim crashed.
6351 35. If a DNS black list lookup returned more than one A record, Exim ignored
6352 all but the first. It now scans all returned addresses if a particular IP
6353 value is being sought. In this situation, the contents of the
6354 $dnslist_value variable are a list of all the addresses, separated by a
6357 36. Tightened up the rules for host name lookups using reverse DNS. Exim used
6358 to accept a host name and all its aliases if the forward lookup for any of
6359 them yielded the IP address of the incoming connection. Now it accepts only
6360 those names whose forward lookup yields the correct IP address. Any other
6361 names are discarded. This closes a loophole whereby a rogue DNS
6362 administrator could create reverse DNS records to break through a
6363 wildcarded host restriction in an ACL.
6365 37. If a user filter or a system filter that ran in a subprocess used any of
6366 the numerical variables ($1, $2 etc), or $thisaddress, in a pipe command,
6367 the wrong values were passed to the pipe command ($thisaddress had the
6368 value of $0, $0 had the value of $1, etc). This bug was introduced by
6369 change 4.11/101, and not discovered because I wrote an inadequate test. :-(
6371 38. Improved the line breaking for long SMTP error messages from ACLs.
6372 Previously, if there was no break point between 40 and 75 characters, Exim
6373 left the rest of the message alone. Two changes have been made: (a) I've
6374 reduced the minimum length to 35 characters; (b) if it can't find a break
6375 point between 35 and 75 characters, it looks ahead and uses the first one
6376 that it finds. This may give the occasional overlong line, but at least the
6377 remaining text gets split now.
6379 39. Change 82 of 4.11 was unimaginative. It assumed the limit on the number of
6380 file descriptors might be low, and that setting 1000 would always raise it.
6381 It turns out that in some environments, the limit is already over 1000 and
6382 that lowering it causes trouble. So now Exim takes care not to decrease it.
6384 40. When delivering a message, the value of $return_path is set to $sender_
6385 address at the start of routing (routers may change the value). By an
6386 oversight, this default was not being set up when an address was tested by
6387 -bt or -bv, which affected the outcome if any router or filter referred to
6390 41. The idea of the "warn" ACL verb is that it adds a header or writes to the
6391 log only when "message" or "log_message" are set. However, if one of the
6392 conditions was an address verification, or a call to a nested ACL, the
6393 messages generated by the underlying test were being passed through. This
6394 no longer happens. The underlying message is available in $acl_verify_
6395 message for both "message" and "log_message" expansions, so it can be
6396 passed through if needed.
6398 42. Added RFC 2047 interpretation of header lines for $h_ expansions, with a
6399 new expansion $bh_ to give the encoded byte string without charset
6400 translation. Translation happens only if iconv() is available; HAVE_ICONV
6401 indicates this at build time. HEADERS_CHARSET gives the charset to
6402 translate to; headers_charset can change it in the configuration, and
6403 "headers charset" can change it in an individual filter file.
6405 43. Now that we have a default RFC 2047 charset (see above), the code in Exim
6406 that creates RFC 2047 encoded "words" labels them as that charset instead
6407 of always using iso-8859-1. The cases are (i) the explicit ${rfc2047:
6408 expansion operator; (ii) when Exim creates a From: line for a local
6409 message; (iii) when a header line is rewritten to include a "phrase" part.
6411 44. Nasty bug in exiqsumm: the regex to skip already-delivered addresses was
6412 buggy, causing it to skip the first lines of messages whose message ID
6413 ended in 'D'. This would not have bitten before Exim release 4.14, because
6414 message IDs were unlikely to end in 'D' before then. The effect was to have
6415 incorrect size information for certain domains.
6417 45. #include "config.h" was missing at the start of the crypt16.c module. This
6418 caused trouble on Tru64 (aka OSF1) systems, because HAVE_CRYPT16 was not
6421 46. If there was a timeout during a "random" callout check, Exim treated it as
6422 a failure of the random address, and carried on sending RSET and the real
6423 address. If the delay was just some slowness somewhere, the response to the
6424 original RCPT would be taken as a response to RSET and so on, causing
6425 mayhem of various kinds.
6427 47. Change 50 for 4.20 was a heap of junk. I don't know what I was thinking
6428 when I implemented it. It didn't allow for the fact that some option values
6429 may legitimatetly be negative (e.g. size_addition), and it didn't even do
6430 the right test for positive values.
6432 48. Domain names in DNS records are case-independent. Exim always looks them up
6433 in lower case. Some resolvers return domain names in exactly the case they
6434 appear in the zone file, that is, they may contain uppercase letters. Not
6435 all resolvers do this - some return always lower case. Exim was treating a
6436 change of case by a resolver as a change of domain, similar to a widening
6437 of a domain abbreviation. This triggered its re-routing code and so it was
6438 trying to route what was effectively the same domain again. This normally
6439 caused routing to fail (because the router wouldn't handle the domain
6440 twice). Now Exim checks for this case specially, and just changes the
6441 casing of the domain that it ultimately uses when it transmits the message
6444 49. Added Sieve (RFC 3028) support, courtesy of Michael Haardt's contributed
6447 50. If a filter generated a file delivery with a non-absolute name (possible if
6448 no home directory exists for the router), the forbid_file option was not
6451 51. Added '&' feature to dnslists, to provide bit mask matching in addition to
6452 the existing equality matching.
6454 52. Exim was using ints instead of ino_t variables in some places where it was
6455 dealing with inode numbers.
6457 53. If TMPDIR is defined in Local/Makefile (default in src/EDITME is
6458 TMPDIR="/tmp"), Exim checks for the presence of an environment variable
6459 called TMPDIR, and if it finds it is different, it changes its value.
6461 54. The smtp_printf() function is now made available to local_scan() so
6462 additional output lines can be written before returning. There is also an
6463 smtp_fflush() function to enable the detection of a dropped connection.
6464 The variables smtp_input and smtp_batched_input are exported to
6467 55. Changed the default runtime configuration: the message "Unknown user"
6468 has been removed from the ACL, and instead placed on the localuser router,
6469 using the cannot_route_message feature. This means that any verification
6470 failures that generate their own messages won't get overridden. Similarly,
6471 the "Unrouteable address" message that was in the ACL for unverifiable
6472 relay addresses has also been removed.
6474 56. Added hosts_avoid_esmtp to the smtp transport.
6476 57. The exicyclog script was not checking for the esoteric option
6477 CONFIGURE_FILE_USE_EUID in the Local/Makefile. It now does this, but it
6478 will work only if exicyclog is run under the appropriate euid.
6480 58. Following a discussion on the list, the rules by which Exim recognises line
6481 endings on incoming messages have been changed. The -dropcr and drop_cr
6482 options are now no-ops, retained only for backwards compatibility. The
6483 following line terminators are recognized: LF CRLF CR. However, special
6484 processing applies to CR:
6486 (i) The sequence CR . CR does *not* terminate an incoming SMTP message,
6487 nor a local message in the state where . is a terminator.
6489 (ii) If a bare CR is encountered in a header line, an extra space is added
6490 after the line terminator so as not to end the header. The reasoning
6491 behind this is that bare CRs in header lines are most likely either
6492 to be mistakes, or people trying to play silly games.
6494 59. The size of a message, as listed by "-bp" or in the Exim monitor window,
6495 was being incorrectly given as 18 bytes larger than it should have been.
6496 This is a VOB (very old bug).
6498 60. This may never have affected anything current, but just in case it has:
6499 When the local host is found other than at the start of a list of hosts,
6500 the local host, those with the same MX, and any that follow, are discarded.
6501 When the list in question was part of a longer list of hosts, the following
6502 hosts (not currently being processed) were also being discarded. This no
6503 longer happens. I'm not sure if this situation could ever has previously
6506 61. Added the "/MX" feature to lists of hosts in the manualroute and query
6509 62. Whenever Exim generates a new message, it now adds an Auto-Submitted:
6510 header. This is something that is recommended in a new Internet Draft, and
6511 is something that is documented as being done by Sendmail. There are two
6512 possible values. For messages generated by the autoreply transport, Exim
6515 Auto-Submitted: auto-replied
6517 whereas for all other generated messages (e.g. bounces) it adds
6519 Auto-Submitted: auto-generated
6521 63. The "personal" condition in filters now includes a test for the
6522 Auto-Submitted: header. If it contains the string "auto-" the message it
6523 not considered personal.
6525 64. Added rcpt_include_affixes as a generic transport option.
6527 65. Added queue_only_override (default true).
6529 66. Added the syslog_duplication option.
6531 67. If what should have been the first header line of a message consisted of
6532 a space followed by a colon, Exim was mis-interpreting it as a header line.
6533 It isn't of course - it is syntactically invalid and should therefore be
6534 treated as the start of the message body. The misbehaviour could have
6535 caused a number of strange effects, including loss of data in subsequent
6536 header lines, and spool format errors.
6538 68. Formerly, the AUTH parameter on a MAIL command was trusted only if the
6539 client host had authenticated. This control can now be exercised by an ACL
6540 for more flexibility.
6542 69. By default, callouts do not happen when testing with -bh. There is now a
6543 variant, -bhc, which does actually run the callout code, including
6544 consulting and updating the callout cache.
6546 70. Added support for saslauthd authentication, courtesy of Alexander
6549 71. If statvfs() failed on the spool or log directories while checking their
6550 size for availability, Exim confusingly gave the error "space shortage".
6551 Furthermore, in debugging mode it crashed with a floating point exception.
6552 These checks are done if check_{spool,log}_{space,inodes} are set, and when
6553 an SMTP message arrives with SIZE= on the MAIL command. As this is a really
6554 serious problem, Exim now writes to the main and panic logs when this
6555 happens, with details of the failure. It then refuses to accept the
6556 incoming message, giving the message "spool directory problem" or "log
6557 directory problem" with a 421 code for SMTP messages.
6559 72. When Exim is about to re-exec itself, it ensures that the file descriptors
6560 0, 1, and 2 exist, because some OS complain for execs without them (see
6561 ChangeLog 4.05/30). If necessary, Exim opens /dev/null to use for these
6562 descriptors. However, the code omitted to check that the open succeeded,
6563 causing mysterious errors if for some reason the permissions on /dev/null
6564 got screwed. Now Exim writes a message to the main and panic logs, and
6565 bombs out if it can't open /dev/null.
6567 73. Re-vamped the way daemon_smtp_port, local_interfaces, and -oX work and
6568 interact so that it is all more flexible. It is supposed to remain
6569 backwards compatible. Also added extra_local_interfaces.
6571 74. Invalid data sent to a SPA (NTLM) server authenticator could cause the code
6572 to bomb out with an assertion failure - to the client this appears as a
6573 connection drop. This problem occurs in the part of the code that was taken
6574 from the Samba project. Fortunately, the assertion is in a very simple
6575 function, so I have fixed this by reproducing the function inline in the
6576 one place where it is called, and arranging for authentication to fail
6577 instead of killing the process with assert().
6579 75. The SPA client code was not working when the server requested OEM rather
6580 than Unicode encoding.
6582 76. Added code to make require_files with a specific uid setting more usable in
6583 the case where statting the file as root fails - usually a non-root-mounted
6584 NFS file system. When this happens and the failure is EACCES, Exim now
6585 forks a subprocess and does the per-uid checking as the relevant uid.
6587 77. Added process_log_path.
6589 78. If log_file_path was not explicitly set, a setting of check_log_space or
6590 check_log_inodes was ignored.
6592 79. If a space check for the spool or log partitions fails, the incident is now
6593 logged. Of course, in the latter case the data may get lost...
6595 80. Added the %p formatting code to string_format() so that it can be used to
6596 print addresses in debug_print(). Adjusted all the address printing in the
6597 debugging in store.c to use %p rather than %d.
6599 81. There was a concern that a line of code in smtp_in.c could overflow a
6600 buffer if a HELO/EHLO command was given followed by 500 or so spaces. As
6601 initially expressed, the concern was not well-founded, because trailing
6602 spaces are removed early. However, if the trailing spaces were followed by
6603 a NULL, they did not get removed, so the overflow was possible. Two fixes
6606 (a) I re-wrote the offending code in a cleaner fashion.
6607 (b) If an incoming SMTP command contains a NULL character, it is rejected
6610 82. When Exim changes uid/gid to the Exim user at daemon start time, it now
6611 runs initgroups(), so that if the Exim user is in any additional groups,
6612 they will be used during message reception.
6618 The change log for 4.20 and earlier releases has been archived.
git://git.exim.org / exim.git / blob